Phase I/II trial of a long peptide vaccine (LPV7) plus toll-like receptor (TLR) agonists with or without incomplete Freund’s adjuvant (IFA) for resected high-risk melanoma

Background We performed a clinical trial to evaluate safety and immunogenicity of a novel long peptide vaccine administered in combinations of incomplete Freund’s adjuvant (IFA) and agonists for TLR3 (polyICLC) and TLR7/8 (resiquimod). We hypothesized that T cell responses to minimal epitope peptides (MEPs) within the long peptides would be enhanced compared with prior vaccines with MEP themselves and that T cell responses would be enhanced with TLR agonists, compared with IFA alone. Methods Participants with resected stage IIB-IV melanoma were vaccinated with seven long melanoma peptides (LPV7) from tyrosinase, gp100, MAGE-A1, MAGE-A10, and NY-ESO-1, each containing a known MEP for CD8+ T cells, plus a tetanus helper peptide (Tet) restricted by Class II MHC. Enrollment was guided by an adaptive design to one of seven adjuvant combinations. Vaccines were administered at weeks 1, 2, 3, 6, 9, 12 at rotating injection sites. T cell and IgG antibody (Ab) responses were measured with IFN-gamma ELIspot assay ex vivo and ELISA, respectively. Results Fifty eligible participants were assigned to seven study groups, with highest enrollment on arm E (LPV7+Tet+IFA+polyICLC). There was one dose-limiting toxicity (DLT) in Group E (grade 3 injection site reaction, 6% DLT rate). All other treatment-related adverse events were grades 1–2. The CD8+ T cell immune response rate (IRR) to MEPs was 18%, less than in prior studies using MEP vaccines in IFA. The CD8+ T cell IRR trended higher for IFA-containing adjuvants (24%) than adjuvants containing only TLR agonists (6%). Overall T cell IRR to full-length LPV7 was 30%; CD4+ T cell IRR to Tet was 40%, and serum Ab IRR to LPV7 was 84%. These IRRs also trended higher for IFA-containing adjuvants (36% vs 18%, 48% vs 24%, and 97% vs 60%, respectively). Conclusions The LPV7 vaccine is safe with each of seven adjuvant strategies and induced T cell responses to CD8 MEPs ex vivo in a subset of patients but did not enhance IRRs compared with prior vaccines using short peptides. Immunogenicity was supported more by IFA than by TLR agonists alone and may be enhanced by polyICLC plus IFA. Trial registration number NCT02126579.


Protocol Precis
Title: Phase I/II trial of a long peptide vaccine (LPV7) + TLR agonists for resected stage IIB-IV melanoma Objectives: Part 1 (completed) (1) To test safety of vaccination w ith a mixture of long peptides for patients w ith high-risk melanoma, in each of 7 adjuvant preparations, including TLR agonists and/or incomplete Freund's adjuvant (IFA ) (2) To estimate the immunogenicity of 7 long peptides in each of 7 adjuvant preparations Exploratory objective: (3) To evaluate the molecular and cellular changes at the vaccine site and systemically Part 2 (4) To assess safety of vaccination of the optimal combination identified in Part 1 w hen administered in the same skin site for all 6 vaccines (same site vaccination).
(5) To estimate the immunogenicity of the optimal combination identified in Part 1 w hen administered in the same skin site for all 6 vaccines (same site vaccination).
Design: This is an open-label, randomized, phase I/II study, using an adaptive design, of the safety and immunogenicity of a novel melanoma vaccination approach using long peptides plus TLR agonists, along w ith characterization of cellular and molecular events at the cutaneous sites of immunization and systemically. This trial involves tw o parts, w hich differ in the location of vaccine injections. For part 1, each vaccine was administered in a skin site that differs on each vaccine date (rotating vaccine sites). For part 2, each vaccine w ill be administered in the same skin site for all 6 vaccines (same site vaccination). induce helper T cell responses themselves (unpublished data), and prior data that adding the tetanus helper peptide did not add significantly to immunologic or clinical outcome of a peptide vaccine designed to induce CD8 T cells 1 . Regimen: Patients w ill be registered on this phase I/II trial and w ill be vaccinated w ith LPV7 at 300 mcg peptide/dose every w eek x 3 then every 3 w eeks x 3.
For Part 1, the adjuvants are described, and w ill be administered to patients in 7 study groups as show n in Table 1A. This incorporates an adaptive design w ith sequential initiation of accrual to study Zones w ith increasing numbers of adjuvants (Arms A -C, then D-F, then G). Details are in Section 9.0 and Appendix 1 (X-page).
For Part 2, the adjuvants w ill be administered to patients in 1 study group as show n in Table 1B.
Vaccine site biopsies: In each patient, three 4-mm punch biopsies w ill be performed of the skin at vaccine sites at tw o time points: 1 w eek after vaccine #1 (day 8) and one w eek after vaccine #3 (day 22). As negative controls, 3 punch biopsies w ill be obtained of clinically normal skin (not at the vaccine site) in the first 6 patients in Part 1 at the same time points (days 8 and 22). Details are provided in section 5.4.6.2.
Leadership: This trial w ill be performed at tw o institutions in patients w ith high-risk resected stage IIB-IV melanoma. The lead institution for this trial w ill be the University of Virginia, and the PI w ill be Dr. Craig Slingluff. The institutional PI at M.D. Anderson Cancer Center w ill be Dr. Patrick Hw u. Part 2 w ill be conducted at the University of Virginia only.
Population: Criteria for inclusion include:  age 18 years or older  expression of HLA-A1, -A2, -A3, -B35, or -B51  histologically or cytologically proven Stage IIB-IV melanoma rendered clinically free of disease by surgery, other therapy, or spontaneous remission. Small (<1cm) findings that are equivocal on radiologic or clinical imaging w ill not be a basis for exclusion.  ECOG performance status 0-1 (Appendix 3)  Adequate hepatic and renal function  Weight at least 50 kg (110 lbs) NOTE: Patients w ill be excluded if they are pregnant or are immunosuppressed, or have previously received vaccines containing an incomplete Freund's adjuvant.
Accrual: For Part 1, the maximum accrual to the study is estimated from the simulations to be approximately 52 patients. Adjusting for 10% ineligibility/dropout/w ithdraw al rate, maximum total target accrual is estimated at 58 patients. For Part 2, target accrual is estimated to be 16 eligible patients. Adjusting for 10% ineligibility/dropout/w ithdraw al rate, maximum total accrual for part 2 is estimated at 18 patients. Clinical outcome: The primary goals are to assess safety and immunogenicity. The study is not pow ered for differences in clinical outcome, but w e w ill track disease-free survival, and overall survival. Significant differences in clinical outcome are not anticipated among the study arms, but trends in clinical outcome w ill be assessed in the context of the immunologic data to support decisions about the preferred vaccination strategy to test further in combination w ith other immune therapies.
Vaccine composition: The peptides described in Table 2 w ill be used in the vaccines.  Safety and toxicity follow ing vaccination w ith 7 long peptides in melanoma patients w ith and w ithout TLR agonists  Levels of peptide-reactive CD8 + T cells in the peripheral blood (peak and durable at 26 w eeks) and vaccine site Secondary:  CD4 + T cell responses to peptides in the vaccine, and their function. Exploratory:  To analyze cellular infiltrates in the VSME  To identify cytokine profiles of the VSME as a function of TLR signaling  To identify toll-like receptor signaling in the VSME  To understand regulatory processes in the replicate immunization sites o Regulatory T cells (CD4 + CD25 hi FoxP3 + ) o Myeloid-derived suppressor cells  Overall Survival and disease-free survival

INTRODUCTION
Abstract Identification of minimal peptide epitopes for melanoma reactive T cells in the early 1990s has provided useful tools for evaluating T cell responses to melanoma; how ever, the immune responses to vaccines using short peptides are often transient and of low magnitude, and they have had limited clinical activity. Recent w ork suggests that vaccination w ith longer (30-mer) peptides that encompass these short peptides may be a more effective strategy and has been associated w ith clinical regressions of squamous vulvar neoplasia 1 . The most common adjuvant for peptide vaccines for melanoma has been a form of incomplete Freund's adjuvant (IFA). How ever, Toll-like receptor (TLR) agonists offer the potential to improve the magnitude and persistence of antitumor T cell responses, and findings from a few clinical trials support this potential [2][3][4] . How ever, there has been very limited characterization of the molecular and cellular effects of TLR agonists at the vaccine site microenvironment (VSME). A randomized trial of high-dose interleukin-2 (IL2) vs. IL2 + peptide vaccine show ed improved response rate and progression free survival w ith addition of the peptide vaccine 5 . Other preclinical studies show improved tumor control by adding vaccines to adoptive T cell therapy and to blockade of immunologic checkpoints. An optimized vaccine strategy, relevant across the large range of HLA alleles is needed.
The proposal is for a clinical trial, performed in a collaborative multicenter setting involving 2 academic institutions. This trial w ill accrue approximately 52 eligible patients (Part 1) and approximately eligible 18 patients (Part 2) in a phase I/II design w ith a long peptide vaccine (LPV7) in one of 7 adjuvant strategies of IFA alone, TLR agonist(s) alone, or a combination of IFA plus TLR agonist(s). Goals of this trial w ill include a safety assessment and measures of immunogenicity, and characterization of the cellular and molecular events induced by each TLR agonist in the VSME. This w ill be a first-inhumans evaluation of the safety and immunogenicity of 7 long 30-mer peptides (LPV7) for melanoma, w hich incorporate 7 w ell-characterized minimal epitopes restricted by HLA-A1, A2, and A3 (approximately 80% of melanoma patients) and a w ell-characterized epitope restricted by HLA-B35/B51. We propose to test these peptides in IFA alone, an adjuvant combining IFA w ith each of the follow ing polyICLC/Hiltonol (TLR3 agonist), resiquimod (TLR7/8 agonist), and both polyICLC and resiquimod, to be administered locally at the site of vaccination.

Study Rationale
Cancer immunotherapy for solid tumors is coming of age, w ith FDA -approved immunotherapeutics now available for the treatment of multiple tumor types. In melanoma, checkpoint inhibitors specific for CTLA-4 and PD-1 have been show n to induce durable clinical regressions and improve overall survival. There is excitement about the grow ing armamentarium of systemic immunotherapeutics, w hose effects are mediated predominantly by T lymphocytes. How ever, despite the effectiveness of those therapies, checkpoint inhibitors fail in about 70-80% of patients, and these patients often succumb to their disease. There is a need for new combination approaches that build on the demonstrated clinical value of immune therapy.
Cancer vaccines inducing antigen-specific T cell responses are emerging as a component of combination immunotherapy. In the past few years, a cancer vaccine has been approved for prostate cancer, based on tw o randomized trials show ing improved BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) survival and a randomized prospective trial show ed that adding a peptide vaccine to highdose IL-2 significantly prolonged progression-free survival (PFS) w hen compared to IL-2 alone. 83;5 Thus, after several decades of development and optimization, there is now evidence that cancer vaccines may improve clinical outcomes, in particular in combination w ith other therapy.
There is a need to improve defined antigens for melanoma vaccines. Peptide vaccines for melanoma offer the promise of inducing T cells reactive to w ell-characterized tumor antigens and also enabling assessment of effectiveness of the vaccinations by monitoring antigenspecific T cell responses. How ever, clinical experience w ith peptide vaccines has been mixed. On one hand, w e and others have found that selected peptides can induce circulating T cell responses in a majority of patients, 8;9 and that vaccination w ith a mixture of peptides can be immunogenic in up to 100% of patients. 8 In some trials, circulating immune responses have correlated w ith clinical outcome 8;10 , but this has not been the rule. Also, the magnitude of T cell responses sometimes is substantial, w ith 1-5% of circulating CD8 T cells reactive to a single antigen 4;11;12 ; how ever, responses in most patients represent T cell percentages that are 1-2 orders of magnitude low er, w hich may or may not be adequate for clinical benefit. The T cell responses to vaccines may be very durable, for months or years, but are at least as likely to be transient, sometimes declining even w hile still receiving vaccines. 13 Clinically, there have been durable clinical responses in some patients receiving melanoma vaccines, suggesting that there is clinical activity. 14 How ever, the overall clinical response rates are only about 3-5%. 15 Thus, vaccines are not optimized, and both the antigens and the adjuvants may be improved.
Long peptides. Recent w ork w ith longer (30-mer) peptides that encompass short minimal epitope peptides suggests that these longer peptides may be more effective immunogens than the minimal peptides. The extra length contributes to a tertiary structure that may protect from peptidases, and they are too long to be presented directly on MHC; so intracellular processing is required. Importantly, this further ensures that the peptides are presented just on professional APC. A vaccine using long peptides for squamous vulvar neoplasia has been associated w ith high rates of clinical regressions. 1 This proposal evaluates long peptides as immunogens for melanoma vaccines. It is for a first-in-human evaluation of safety and immunogenicity of 7 long peptides for melanoma, w hich incorporate w ell-characterized minimal epitopes restricted by HLA -A1, A2, and A3 and HLA-B35/51. Adjuvants for cancer vaccines. Some data have challenged the effectiveness of the current formulation of incomplete Freund's adjuvant (IFA) w ith a peptide vaccine, especially for induction of Th1/Tc1 T cell responses, 16 w hile data from our ow n experience supports its adjuvanticity 17 . How ever, data from our ow n experience, and from our collaborators also show that vaccines using short peptides in IFA induce chronic inflammation at the site of vaccination that attracts and retains antigen-specific T cells at the vaccine site and may cause T cell dysfunction 18;19 . Thus, there is a need to optimize adjuvants; there also is reason to expect that IFA w ill function differently w ith long than w ith short peptides, as the epitopes derived from processing the long peptides w ill be presented in the draining nodes, rather at the sites of vaccination.
Toll-like receptor (TLR) agonists offer the potential to improve the magnitude and persistence of antitumor T cell responses 4 ; how ever, most trials of TLR agonists have been limited to use of one TLR agonist formulation, and have not defined the molecular and cellular effects at the vaccine site microenvironment (VSME). In the present study, w e propose to test each of 2 TLR agonists in parallel and in combination, w ith or w ithout IFA, and to evaluate the added value over IFA alone. These include agonists for TLR3, TLR7, and TLR8, to be administered locally at the site of vaccination. BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) Combination immune therapy: This proposal w ill enable selection of an effective immunization strategy for combination w ith other effective systemic immune therapies in future studies. Vaccination w ith a single HLA-A2 restricted peptide in IFA improved response rate and progression-free survival w hen combined w ith HD IL2 in a randomized prospective trial for Stage IV melanoma 5 . Progression-free survival w as the primary endpoint for that study. Despite the fact that the study w as not pow ered to detect differences in survival, the vaccine-treated patients had a strong trend to improvement in overall survival (p = 0.06) 5 . How ever, vaccination w ith HLA-A2 peptides plus IFA alone in another study had no impact on clinical response. 20 The trial that produced the encouraging positive finding in combination w ith HD IL2 needs to repeated both for confirmation of benefit and to evaluate impact on survival in an adequately sized study. Recent concerns w ith the inadequacy of IFA alone as an adjuvant have prevented investigators from moving to a confirmatory trial until a more optimal adjuvant is available.
Other preclinical studies show that adding a vaccine to adoptive T cell therapy increases its therapeutic efficacy. [21][22][23][24] Also, vaccination can synergize w ith PD1/PDL1 blockade to increase immune responses and tumor control. 25 Thus, there is evidence to support the value of combining vaccines plus any of several effective immune therapies. How ever, vaccines are not yet optimized. The present application w ill test a vaccine comprised of multiple long peptides and multiple adjuvants. The use of peptides restricted by HLA -A1, A2, A3, and B alleles has a significant advantage for speeding accrual for a future definitive trial by almost doubling potential eligibility over HLA-A2 restriction alone (more than 80% vs. 45% HLA appropriate).
Vaccine site selection: Preliminary data from analysis of several clinical trials of peptide vaccines at the University of Virginia suggests that patients w ho have been vaccinated at the same skin site each w eek (for at least 3 vaccines, same site vaccination) have had stronger T cell responses than those w ith vaccines administered to a different skin site w ith each vaccine (rotating vaccine sites). How ever, those data w ere based on retrospective assessments among trials w ith varied vaccine adjuvants and over extending time periods. Also, the differences w ere not definitive enough to be assured that site makes a difference in outcome, We hypothesize that same site vaccination w ith LPV7 w ill be safe and w ill result in stronger CD4 + and CD8 + T cell responses to those peptides. Preliminary data from Part 2 w ill assess w hether there may be a basis for exploring this question in a definitive randomized trial, and w ill provide a basis for pow ering such a study.

Antigen-Specific Immune Responses Induced by Peptide Vaccines
We have substantial experience in studies w ith multipeptide vaccines, and w ith high antigen-specific immune response rates in the peripheral blood and in vaccine-draining nodes. 8;9;12 These provide a strong foundation for evaluation of the multipeptide vaccines proposed in this application, and for the immunologic analyses.

Melanoma Peptides Restricted by MHC Class I Molecules and
Incorporated in the prior vaccine trials, that are selected for use in the LPV7 mixture Antigens representing the majority of melanoma-associated antigens identified thus far can be divided into tw o groups. The first group consists of melanocytic differentiation proteins (MDP); such antigens are only expressed on cells of the melanocytic lineage and include MART-1/MelanA, gp100/Pmel17, tyrosinase, TRP-1/gp75 and TRP-2 proteins. The second group consists of cancer-testis antigens (CTA) such as MAGE proteins, LAGE-1, and NY-ESO-1. These proteins are encoded by genes expressed in several tumors of different BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) histologic types, but not in normal tissues, other than testis and placenta. Peptides derived from these proteins are recognized by CTL in the context of various HLA alleles.
Melanocytic differentiation proteins: Gp100 is an enzyme involved in melanin synthesis. Multiple epitopes restricted by Class I MHC molecules have been identified from this protein, including peptides restricted by HLA-A2, and HLA-A3 [26][27][28] , and w e chose to incorporate in the present proposal long peptides that represent the follow ing short epitopes: gp100209-217-2M IMDQVPFSV (substitution of M for T at position 210; HLA -A2), and ALLAVGATK (HLA-A317-25). The parent gp100209-217 epitope (ITDQVPFSV) has been modified to incorporate a methionine residue at position 210 of the peptide (IMDQVPFSV). This modification increases the binding affinity of the IMDQVPFSV peptide for HLA -A2 and increases its immunogenicity in vivo w hen compared to the native peptide 29;30 . Another gp100 derived epitope, ALLAVGATK, is also naturally processed and presented by melanoma cells, and holds promise for induction of HLA -A3 restricted responses in vivo 28 . CD8 + T cell responses to the ALLAVGATK peptide have been detected in multiple prior melanoma vaccine trials 12;31-34 . T-cell lines cultured from lymphocytes infiltrating human melanomas, w hen administered adoptively to participants w ith metastatic melanoma, have induced partial or complete remissions in a large subset of participants. Clinical response to this therapy has been reported to be increased w hen the T-cells used for therapy recognize peptides derived from Pmel-17/gp100 35 . Thus, there is substantial evidence to support the use of the gp100-derived peptides in the melanoma vaccines.
Tyrosinase is an enzyme involved w ith melanin synthesis. Epitopes derived from tyrosinase have been identified for HLA-A1, HLA-A2, and other class I alleles 26 . In the proposed study, w e are including long peptides incorporating one epitope (DAEKSDICTDEY) restricted by HLA-A1 and one epitope (YMDGTMSQV) restricted by HLA -A2. The parent epitope of tyrosinase240-251, DAEKCDICTDEY, has been modified to incorporate a serine residue at position 244 of the peptide (DAEKSDICTDEY). This modification prevents disulfide bond formation w ithin the peptide, but does not interfere w ith HLA -A1 binding or w ith T cell recognition 36 . In phase II clinical trials conducted by the UVA HITC (UVA -Mel31, Mel36, Mel39, Mel43, and Mel44), immunological responses against the DAEKSDICTDEY peptide w ere detected as the dominant peptide target for participants w ho expressed HLA-A1 and w ho w ere immunized w ith a synthetic peptide mixture containing the DAEKSDICTDEY peptide 12;31-34 . CTL cultured from a vaccine draining node (sentinel immunized node, SIN) after vaccination w ith DAEKSDICTDEY w ere capable of lysing HLA-A1 + tumor cells naturally expressing tyrosinase 37 . The naturally occurring HLA-A2 restricted epitope from tyrosinase, tyrosinase369-377 YMDGTMSQV, contains a post-translational modification of an asparagine to aspartic acid at residue 371 38 . In phase II clinical trials conducted by the UVA HITC, immunological responses to the YMDGTMSQV peptide have been detected frequently in stage III and IV melanoma participants w ho expressed HLA -A2, and w ere immunized w ith a synthetic peptide mixture containing this peptide 12;31-34 .
MAGE-A1 and MAGE-A10 are members of the MAGE gene family and are expressed in the testes, as w ell as in several different tumor types such as melanoma, breast, prostate, esophagus, colon, and lung 26 . One epitope derived from MAGE-A1 (SLFRAVITK) is included in the proposed study. The MAGE-A196-104 epitope (SLFRAVITK) w as identified by Chaux et al. 39 , and w as first tested in humans in UVA-Mel39, in w hich immunogenicity and safety have been demonstrated 33 . It is a highly immunogenic peptide in the Mel39, Mel43, and Mel44 trials 8;12;40 .
The MAGE-A10254-262 epitope (GLYDGMEHL) w as identified by Huang et al., and w as show n to be naturally processed and presented by HLA -A2 + /MAGE-A10 + melanoma cell lines. 41 In a report by Valmori et al., CTL responses to this epitope w ere readily detected in tw o-thirds of participants w ith melanoma w hose tumors tested positive for the expression of MAGE-A10. 42 How ever, the ability of MAGE-A10 synthetic peptide to stimulate immunologic responses in humans had not been tested until UVA -Mel39, in w hich immunogenicity and safety have been demonstrated and this w as found to be very immunogenic 33 . Follow up studies further support immunogenicity of this MAGE-A10 peptide 12;34 .
The cancer testis antigen NY-ESO-1 has been extensively studied in human clinical trials, including peptide, protein, and DNA 2;43-46 . Recently, a set of 4 overlapping long (30-mer) peptides have been tested for safety and immunogenicity in patients w ith NY -ESO-1+ ovarian cancer, using either IFA alone, or IFA plus polyICLC (1.4 mg) 2 . That study incorporated the NY-ESO-1 peptide to be used in the current trial (NY-ESO-179-108; CD4+ and CD8+ T cell responses w ere induced to a mixture of peptides covering the NY -ESO-1 sequence over residues 79-130, in a majority of evaluated patients.
For the present study, the 7 peptides have been selected for use as long peptides. Their sequences are listed in Table 2 in the protocol précis. The reason for selecting six of them is that their minimal epitopes are among the most immunogenic in the set of 12 peptides included in our prior trials 12;33;34 : DAEKSDICTDEY, YMDGTMSQV, IMDQVPFSV, GLYDGMEHL, ALLAVGATK, SLFRAVITK, and they have been administered safely 12;33;34 . We had also selected the peptide EVDPIGHLY from those prior trials, but the long peptide encompassing EVDPGIHLY could not be synthesized w ith adequate purity; the remaining 6 peptides w ere selected for use.
The NY-ESO-179-108 peptide w as included based on the encouraging immunogenicity and safety findings referred to above w ith four NY -ESO-1 peptides 2 . Among the four peptides used in that trial, this one w as available in adequate quantity as GMP-grade preparation.
Selection and synthesis of long peptides encompassing the short peptide antigens: The NY -ESO-179-108 peptide incorporates a defined nonamer epitope restricted by HLA-B35 and HLA-B51; the sequence w as one of 4 previously used as the 30-mer peptide in a phase I trial. For the present study, the 7 long peptides selected for use have been designed to incorporate the minimal epitope sequences above, plus extension at the N and C terminal ends, w ith attention to avoid N-terminal glutamic acid or glutamine residues, to minimize cysteine residues, and to maximize incorporation of polar residues to aid in aqueous solubility. Their sequences are listed in Table 2 in the protocol précis.

Montanide ISA-51 as a Vaccine Adjuvant.
Montanide adjuvant has been effective at inducing immune responses against murine viral antigens w hen administered w ith a synthetic peptide epitope 47;48 . Montanide ISA-51 has also been used w ith peptides plus GM-CSF-in-adjuvant vaccine useful for enhancing both cellular and humoral immunity. 49 The product consists of a mineral oil base comparable to incomplete Freund's adjuvant. How ever, the Arlacel A emulsifying agent of incomplete Freund's, w hich has caused reactions in the past, has been replaced w ith a purified mannoside monooleate called "montanide", w hich appears to be safer. The UVA HITC has sponsored studies w here peptide-based vaccines in Montanide ISA-51 have been safely administered to more than 500 participants.
Immunological responses against the Dosing and preparation: It has been used safely in cancer patients, w ith intravenous doses up to 300 mcg/kg. 53 We w ill administer 1 mg (0.5 mL) per vaccine, as used in other trials (e.g.: NCT01008527).

Resiquimod: TLR7 and 8 agonist.
This agent is an agonist for both TLR7 and TLR8. TLR8 activation may reduce regulatory T cells. 54 It has been safe w hen administered topically as a 0.05% gel once a w eek; more frequent administration led to dose-limiting local skin toxicity. 55 Thus, w eekly dosing is selected for this trial. Other human experience w ith resiquimod is that oral dosing at 0.01 mg/kg (700 mcg/70kg) caused increases in serum IFN-alpha w ith associated side effects, but it w as w ell tolerated. 56 It has been used (NCT00821652) safely at 500 mg 0.2% resiquimod gel per application up to 3x/w eek after each vaccine, w ith the dose spread over a 25 cm2 area each time.
Source: Resiquimod is provided by 3M Pharmaceuticals, St. Paul, MN. It is not yet FDA approved, but is available as a clinical grade reagent for topical use. We have approval from Mark Tomai at 3M to provide it for clinical use as a vaccine adjuvant. Dosing and preparation: Multi-dose tubes contain approximately 3 grams of 0.2% resiquimod gel. The dose of resiquimod in 500 mg of 0.2% resiquimod gel is 1000 mcg. Immediately after vaccine injection, measure out approximately 500 mg of 0.2% resiquimod on dosing paper and apply resiquimod gel w ith a gloved finger over a 25 cm 2 area of the skin at the vaccine site. Using this dosing method, the dose range of resiquimod w ill be approximately 400-600 mg. Rub the gel into the skin for 1 minute until it vanishes. Follow ing application, the site may be covered w ith an occlusive dressing, such as Tegaderm, for about 8 hours. Participants are encouraged to w ash the area at 8 hours after application.The tubes may be retained in the investigational pharmacy betw een uses.

Integration of Peptide-Based Vaccines in the Adjuvant Setting
High-dose interferon (IFN-alpha therapy) is approved by the FDA for use in patients w ith resected stage IIB or III melanoma treated in the adjuvant setting. Approval w as based largely on survival improvement observed follow ing a one-year regimen w ith IFN as described in ECOG 1684. 57 How ever, survival improvement is modest and data from a subsequent trial, ECOG 1690, failed to reproduce differences in survival. 58 Data from a third trial, ECOG 1694, reveal a statistically significant, but still modest, survival improvement w ith IFN therapy compared to a ganglioside vaccine. 59 Thus, any patient w ith resected stage IIB or III melanoma should be offered IFN therapy, w hich is routine practice at UVA and MDACC. How ever, the toxicity of high-dose IFN therapy is substantial. Approximately 25-30% of patients w ithdraw from therapy because of toxicity, and approximately 60% of patients require dose reductions during treatment. 57 Ipilimamb 10mg/kg is approved by the FDA for use in the adjuvant setting for patients w ho have undergone complete resection of cutaneous melanoma w ith pathological involvement of regional lymph nodes greater than 1 mm. Approval w as based largely on results from the the Phase 3 EORTC 18071/CA184029 trial. 84 The ipilimumab 10mg/kg treatment yielded statistically significant prolonged distant metastasis-free survival, w ith five-year distant metastasis-free survival rate 9% higher than the placebo. How ever, the toxicty of ipilimumab 10mg/kg is is substantial, w ith approximately 50% of patients w ho started ipilimumab w ithdraw ing as a result of ipilimumab related toxicities.
The experience at UVA is that most patients w ho are candidates for therapy refuse IFN because of concerns w ith toxicity, despite being informed IFN therapy is one of the few FDA approved adjuvant therapies for melanoma and that therapy may provide a survival advantage. In addition, many patients are not eligible for the current approved dose of ipilimumab (10 mg/kg) or decline treatment w ith ipilimumab because of the potential for serious immune-related toxicities. Recent data from the E1609 study presented at ASCO show ed that there w as no difference in disease-free-survival betw een ipilimumab 3 mg/kg and ipilimumab 10 mg/kg (E1609); how ever, ipilimumab 10 mg/kg w as associated w ith greater toxicity. 85 Thus, future approved treatment options may include ipilimumab administered at the low er dose of 3 mg/kg, although even the low er dose carries risk of death. The same may be true for other checkpoint inhibitors that are currently being explored in the adjuvant setting (e.g. PD-1 blockade).

Toxicology
There is no reason to expect direct toxicity of the melanoma peptides; they are not directly cytotoxic in vitro and are not expected to have functional implications. On the other hand, because some of these peptides are identical or similar to a portion of a normal protein, risks of autoimmunity in humans are reasonable to evaluate. Unfortunately, there is no murine system adequately modeling the human immune response to these peptides. The most meaningful evaluation of this peptide vaccine mixture is in patients w ith melanoma. This trial is for participants w ith resected stage IIB/C, III, or IV NED melanoma. These individuals face a high risk (> 40%) of death from melanoma, and the anticipated risk of short-term or long-term toxicity of this vaccine preparation is minimal, w hile the vaccine may delay or decrease the risk of morbidity and mortality due to melanoma in these participants. The potential implications of autoimmunity against cells of melanocytic lineage are illustrated by reported cases of vitiligo occurring coincident w ith regressions of melanoma. 60 Most of these are limited, often occurring in skin surrounding the regressing melanoma, but occasionally occurring systemically. While pathogenesis of this phenomenon can only be hypothesized, it is reasonable to consider this a w orst-case scenario.
The loss of skin and hair pigment can be striking in cases of vitiligo, but is not a cause of morbidity or mortality. Of greater potential concern is the theoretical risk of damage to the retinal pigment epithelium; how ever, visual loss has not been reported either as a complication of successful immunotherapy w ith melanoma-reactive CTL or spontaneous vitiligo. How ever, some ocular toxicity has been reported that resolved w ith steroid treatment. 61 Depigmentation of the retinal pigment epithelium has been observed in a small number of patients vaccinated w ith dendritic cells pulsed w ith MDP-derived peptides; how ever, this change w as asymptomatic and w as not associated w ith loss of visual acuity (personal communication -Frank Haluska). A careful study of the retinal pigment epithelium using monobenzyl ether of hydroquinone to induce pigment cell destruction on a biochemical basis suggests the safety of pigment cell destruction and supports immunotherapy directed against MDP as a strategy for melanoma therapy (personal communication, JM Kirkw ood).
Thus far, 10-15% of participants receiving peptide-based vaccines in the studies at UVA w ere diagnosed w ith definite vitiligo. In all cases, the vitiligo w as asymptomatic. No major visual/ocular toxicity w as reported for participants on these prior trials.  Table 3: The grading of injection site reactions w as modified from CTC v2.0 to CTCAE v3.0, and now to v4.03. CTC v2.0 did not have an ulceration Adverse Event Term, therefore participants w ith ulceration in conjunction w ith an injection site reaction w ere graded as experiencing a Grade 3 injection site reaction (ulceration or necrosis that is severe; operative intervention indicated). In prior UVA HITC studies, an injection site reaction w ith ulceration ≤ 2 cm in maximal diameter w as an expected adverse event in a subset of participants. How ever, an injection site reaction that included ulceration >2 cm in diameter, or an injection site reaction requiring debridement, narcotic analgesics for pain, or surgery w as not an expected toxicity and w as considered a dose-limiting toxicity (DLT) and w as graded as a Grade 3 injection site reaction. UVA-Mel43 used the CTC v2.0 grading system w hereas UVA -Mel44 used the CTCAE v3.0 grading system. In CTCAE v3.0, ulceration is a separate Adverse Event Term; therefore, ulceration and injection site reactions can be graded separately . For this study, participants w ho experience induration at their injection site, both injection site reaction and induration w ill be graded and reported. The data in Table 3 have been corrected to be consistent w ith CTCAE v3.0. The criteria for each of these toxicity categories (fatigue, injection site reaction, and ulceration) have changed from v3.0 to v4.03, but are very similar. Thus, they are probably comparable, but the exact details for assessing them in v4.03 require information that w as not collected at the time those studies w ere performed. In the proposed study, w e w ill grade based on v4.03, and w e do not expect Grade 3 injection site reactions; ulceration and injection site reaction toxicities w ill be expected at Grade 2.

Patient acceptance of biopsies of skin at the Vaccine Site
This study w ill include three punch biopsies 4mm in diameter of the vaccine site, on tw o dates. We have previously performed much larger excisions of skin and subcutaneous tissue approximately 2 x 4-6 cm in size. 50 There w ere questions, prior to that study, of w hether that size excision w ould be too large to expect participants to tolerate or to accept it. As a surgeon, the Principal Investigator (PI) routinely performs excisions 2-4 cm w ide (and about 6-12 cm long) in patients w hen they have w ide excisions of melanoma, and these can reliably be closed primarily, w ithout a skin graft, w hen taken from the upper thigh and upper arm. Thus, excisions of a 2x4 cm skin section (regardless of depth) can be closed primarily w ith minimal morbidity on those locations w here the vaccines w ould be given.
Also, w e surveyed a convenience sample of 52 of our patients during clinic visits for their w illingness (on a scale of 1 to 9) to participate in a clinical trial of a melanoma vaccine w ith or w ithout a skin biopsy. Forty-three (43) of the 52 patients (83%), answ ered some of the questions, and 37 (71%) answ ered all questions. Findings are summarized in Table 4 below and have been published 62 . Willingness to enter such a trial w as diminished by any biopsy requirement, but interest in accrual remained high even w ith a biopsy. Interestingly the size of the biopsy had a much smaller impact than the fact that any biopsy w as required. In prior trials, w e have included excision of a vaccine-draining lymph node (sentinel immunized node), w hich requires a 3 cm incision, and w e have done over 150 of such biopsies on vaccine patients. 31-33;37 Based on that questionnaire, patients w ere expected to be about 70-80% as likely to enroll on a trial w ith a 6 cm excision vs. a 3 cm excision as w ithout a biopsy. The actual accrual matched closely the predicted accrual. 62 The present proposal is to remove much less skin that w as done in that prior protocol (Mel48). Thus, w e anticipate rapid accrual and high patient acceptance. We typically have accrued 40-60 patients per year on trials w ith sentinel immunized node biopsies; so if w e accrue at even half that rate (20-30 per year), w e expect to meet accrual goals.

Table 4. Findings from vaccine biopsy questionnaire (n = 91)
Willingness to enter vaccine trial with or without skin excision N Likely (5) to highly likely (9) Moderately likely (7) to highly likely (9) Highly likely (score has limited access, an entry room for gow ning and gloving, and positive pressure ventilation and laminar flow hoods. Lot release testing of the final vialed peptides w ill be completed in accord w ith CFR guidelines. Stability testing w ill be conducted regularly, in accord w ith the Chemistry and Manufacturing section of the IND application. Montanide ISA-51 is produced by Seppic, Inc. (Fairfield, NJ and Paris, France). A master drug file for Montanide ISA-51 is filed w ith the FDA. This agent has been used in hundreds of patients in our ow n experience and in thousands of patients w orldw ide.
This study w ill be approved by the Institutional Review Board (IRB) of the University of Virginia and by the IRBs of M.D. Anderson Cancer Center. This protocol w ill be submitted to the FDA as an IND application. Records of all study review and approval documents w ill be kept on file by the Principal Investigator, and are subject to FDA inspection during or after completion of the study. The Institutional Review Board w ill receive notification of study closure w ithin three months of study termination or completion.
This study w ill be conducted in compliance w ith Good Clinical Practices and regulations.

Same site vaccination (Part 2)
The clinical trials group at the University of Virginia has extensive experience in several hundred patients w ith administering peptide vaccines in IFA, w ith or w ithout other immune modulators, at the same skin site for each vaccine. This has been w ell-tolerated 34;77-99 . Later trials w ith rotating vaccine sites w ere performed for several reasons, but w ith the expectation that vaccine site reactions w ould be less prominent; how ever, overall vaccine site reactions have been similar w ith both approaches (Table 5): The LPV7 vaccine has been w ell-tolerated in the patients on Part 1. Same site vaccination w ill be assessed for safety/tolerability and immunogenicity in a pilot cohort in Part 2.

Part 1 (completed)
(1) To assess safety of vaccination w ith a mixture of long peptides for patients w ith highrisk melanoma, in each of 7 adjuvant preparations, including TLR agonists and/or incomplete Freund's adjuvant (IFA) 2.2 Part 2 (4) To assess safety of vaccination of the optimal combination identified in Part 1 w hen administered in the same skin site for all 6 vaccines (same site vaccination).
(5) To estimate the immunogenicity of the optimal combination identified in Part 1 w hen administered in the same skin site for all 6 vaccines (same site vaccination).

Type of Study
This is an open-label, randomized, phase I/II study, using an adaptive design, to test the safety and immunogenicity of a novel melanoma vaccination approach using long peptides plus TLR agonists, along w ith characterization of cellular and molecular events at the cutaneous site of immunization. Regimen: Part 1: Patients w ill be registered on this phase I/II trial and w ill be vaccinated w ith LPV7 at 300 mcg peptide/dose, and tetanus peptide at 200 mcg/dose, every w eek x 3 then every 3 w eeks x 3. The adjuvants are described, and w ill be administered to patients in 7 study groups as show n in Tables 1 and 5. This incorporates an adaptive design w ith sequential initiation of accrual to study Zones w ith increasing numbers of adjuvants (Arms A-C, then D-F, then G) and a 3 w eek interval betw een subjects 1 and 2 of each arm of the study. The design allow s, during that 3 w eek interval, randomization to arms that remain open.
Part 2: Patients w ill be registered on this phase I/II trial and w ill be vaccinated w ith LPV7 at 300 mcg peptide/dose every w eek x 3 then every 3 w eeks x 3. For part 2, the vaccine regimen is selected based on available data and the study decision rules from Part 1 w hich w as determined to be Arm E. For Part 2 accrual w ill be to Study Arm E2. How ever, a modification is made to exclude the tetanus helper peptide, due to (a) lack of availability of that peptide component, (b) the finding that, in part 1, LPV7 peptides do induce helper T cell responses themselves (unpublished data), and prior data that adding the tetanus helper peptide did not add significantly to immunologic or clinical outcome of a peptide vaccine designed to induce CD8 T cells 1 .

PARTICIPANT SELECTION
All questions regarding eligibility should be directed to the Human Immune Therapy Center (HITC) at the University of Virginia.

Inclusion Criteria
4.1.1 Histologically or cytologically proven Stage IIB-IV melanoma (at diagnosis or at the time of recurrence) rendered clinically free of disease by surgery, other therapy, or spontaneous remission w ithin 6 months prior to registration; patients w ith treated brain metastases may be eligible if they meet the criteria in 4.1.4. Small radiologic or clinical findings of an indeterminate nature w ill not be a basis for exclusion, and brain metastases treated in accord w ith section 4.1.4 w ill not be a basis for exclusion.
Staging of cutaneous melanoma w ill be based on the 7 th edition AJCC staging system (Appendix 2) 63 .  The total number of brain metastases ever is less than or equal to 3.  The brain metastases have been completely removed by surgery or have been treated completely by stereotactic radiotherapy. Stereotactic radiotherapy, such as gamma knife, can be used up to 1 w eek prior to study entry. In the absence of concerning clinical findings, repeat scans are not required after stereotactic radiotherapy if the patient enrolls w ithin 8 w eeks of completing the stereotactic therapy.  There has been no evident grow th of any brain metastasis since treatment.  No treated brain metastasis is greater than 2 cm in diameter at the time of protocol entry.
4.1.5 Patients must have at least one axillary and/or inguinal lymph node basin that is intact (no prior excisional biopsy of a node or complete lymph node dissection). w eeks (except as specified in section 4.2.4). Gamma knife or stereotactic radiosurgery may be administered w ithin the prior 4 w eeks, but must not be administered less than one w eek prior to study registration. Patients w ho are currently receiving nitrosoureas or w ho have received this therapy w ithin the preceding 6 w eeks are excluded. 4.2.4 Patients w ho are receiving or have been treated w ith antibody to CTLA4 (e.g. ipilimumab), PD-1, PD-L1, CD137, or CD27 w ithin the prior 12 months. Any patient w ho has had one or more of these therapies greater than 12 months prior and is clinically free of disease and totally recovered from toxicities related to those therapies can be eligible.  vaccination. Women must also not be breast feeding. This is consistent w ith existing standards of practice for vaccine and chemotherapy protocols. 4.2.11 Participants must not have had prior autoimmune disorders requiring cytotoxic or immunosuppressive therapy, or autoimmune disorders w ith visceral involvement. Participants w ith an active autoimmune disorder requiring these therapies are also excluded. The follow ing w ill not be exclusionary:  The presence of laboratory evidence of autoimmune disease (e.g. positive ANA titer) w ithout symptoms  Clinical evidence of vitiligo  Hypothyroidism of any etiology on stable thyroid hormone replacement therapy.  Other forms of depigmenting illness  Mild arthritis requiring NSAID medications 4.2.12 Patients in w hom there is a medical contraindication or potential problem in complying w ith the requirements of the protocol, in the opinion of the investigator.

4.2.13
Patients classified according to the New York Heart Association classification as having Class III or IV heart disease (Appendix 4).

4.2.
14 Patients w ith a body w eight < 110 lbs (50 kg) because of the amount and frequency w ith w hich blood w ill be draw n, and because of the skin biopsies required.

Participant Registration
All participants must sign the consent form prior to determination of eligibility for this study. Participants eligible for interferon therapy w ill complete the interferon education packet and review their answ ers w ith the study team and convince the study team that they answ ered the questions accurately or understand the correct answ ers. Registration information, the eligibility checklist w ith supporting documentation, and the on-study case report form for potential participants at institutions other than UVA must be received by the UVA HITC prior to registration and randomization. All participants w ho meet the inclusion/exclusion criteria may be registered to the study. Registration w ill occur follow ing verification of eligibility by the treating physician or institutional PI.

Randomization
Randomization w ill be discussed w ith participants during the process of informed consent, and informed consent must be documented prior to randomization. Randomization w ill occur after registration and no sooner than 5 days prior to the start of treatment. In part 1, randomization w ill be based on equal allocation among allow able arms unless a w eighted allocation scheme is triggered (section 9.2.1). Randomization w ill not be stratified by BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) institution. The randomization codes are generated by the study statisticians and stored in the Cancer Center Clinical Trials Database. The UVA HITC w ill communicate w ith the outside site to let them know w hich study arm a participant has been randomized to. Participants should receive the first scheduled vaccination or biopsy (if applicable) w ithin 2 w eeks of registration. Part 2 w ill not include randomization.

Management of Participants
This study w ill be conducted on an outpatient basis, w ith participants scheduled to be evaluated on days 1,8,15,22,36,43,57,78, and 85, and w eek 26 (or more often if needed for testing or medical reasons). As needed, there may also be follow -up w ithin 1-3 w eeks after biopsies performed on days 8 or 22 for w ound evaluation. Participants w ill also be follow ed-up at 52 w eeks (1 year) and at 2 years w ith blood draw s for studies of immune response durability. During off treatment follow -up, participants w ill be follow ed yearly for survival and disease status.

Peptide Synthesis and Storage
All peptides are synthesized directly from amino acids, and then are purified by high performance liquid chromatography (HPLC). The identity of the synthetic peptides w ill be confirmed by verifying their mass and amino acid sequences by mass spectrometry. Details of the synthesis, certificates of analysis, and technical summaries are included in the Chemistry and Manufacturing section of the IND application. Each bulk peptide is supplied as lyophilized pow der w ithout excipients and stored at a temperature ≤ -70°C and protected from light.

Preparation of peptide vaccine vials for patient use.
For the 6 peptides prepared at UV A, the bulk peptides are solubilized in aqueous solution, sterile-filtered, and mixed to provide a final concentration of 600 mcg/ml for each of the 6 peptides. The peptide mixture is then aliquoted at 1 ml per vial into sterile borosilicate vials in the HITC clean room. The mixture is slightly cloudy, as the pH of the mixture supports solubility of most of the peptides, but some peptides are partially in suspension at that pH. The 7 th peptide, from NY-ESO-1 is provided as single use vials of lyophilized peptide, at 600 mcg/vial, from the Ludw ig Institute for Cancer Research, prepared under GMP conditions.

Storage of Vialed Peptides
The vials of peptide are stored by the HITC at a temperature ≤ -70°C and protected from light. On the day of use, they are thaw ed at room temperature or in a w ater bath up to 37°C. They should be sw irled gently as they are thaw ed and moved to a refrigerator as soon as they are thaw ed if not being injected immediately. Once thaw ed, the vial(s) must be used for preparation of the vaccine w ithin 24 hours.

Lot Testing
Each lot of peptide vaccine is evaluated as required by the FDA to confirm identity, sterility, general safety, purity, lack of peptide aggregation and is tested for endotoxin. In addition, studies of stability w ill be performed over time. The details of these tests are outlined in Appendix 5.

Labeling
The labeling of vials w ill occur as specified below .
Each vial of lyophilized peptide is labeled w ith the follow ing information: Short name of the product: "LPV6" Product number Proper name of the product: "6 long melanoma peptides" Name and address of the vialing facility: "UVA-HITC" Lot number Date of manufacture (the date of vialing the reconstituted peptides) Serial number Quantity of each peptide per vial: 600 mcg/ml Vial contains no preservative, store at ≤ -70°C "Caution: New Drug -Limited by US Federal law to investigational use"

Dosage and Preparation of Peptide Vaccine Part 1:
The peptides w ill be prepared by combining the LPV 6 mixture of 6 peptides w ith one NY-ESO-1 long peptide. The long melanoma peptides (LPV6) w ill be available as single-use vials of 1 ml volume containing 600 mcg/ml of each peptide. The NY -ESO-1 peptide is provided as lyophilized preparation w ith 600 mcg of peptide in single use vials. The tetanus peptide is vialed as 300 mcg lyophilized peptide per vial. For each vaccine, the full 1 ml of the peptide solution of LPV6 (600 mcg) w ill be w ithdraw n and injected into the vial of lyophilized NY-ESO-1 peptide, creating a solution/suspension of all 7 long peptides, each at 600 mcg/ml. Then 0.75 ml of this suspension/solution w ill be w ithdraw n and injected into one vial of lyophilized tetanus peptide, to create a suspension/solution of 450 mcg of each of the 7 long peptides, and 300 mcg of tetanus peptide in 0.75 ml. Then, the full 0.75 ml of this solution/suspension (450 mcg of all 7 long peptides, plus 300 mcg tetanus peptide) w ill be w ithdraw n into a syringe for injection as specified below for each study arm. Details of making emulsions w ith IFA (Montanide ISA -51) are provided in the Investigator's Brochure and Study Manual.

Part 2:
The peptides w ill be prepared by combining the LPV6 mixture of 6 peptides w ith one NY -ESO-1 long peptide. The long melanoma peptides (LPV6) w ill be available as single-use vials of 1 ml volume containing 600 mcg/ml of each peptide. The NY -ESO-1 peptide is provided as lyophilized preparation w ith 600 mcg of peptide in single use vials. For each vaccine, the full 1 ml of the peptide solution of LPV6 (600 mcg) w ill be w ithdraw n and injected into the vial of lyophilized NY -ESO-1 peptide, creating a solution/suspension of all 7 long peptides, each at 600 mcg/ml. Then 0.75 ml of this solution/suspension (450 mcg of all 7 long peptides) w ill be w ithdraw n into a syringe for injection as specified below for each study arm. Details of making emulsions w ith IFA (Montanide ISA-51) are provided in the Investigator's Brochure and Study Manual.

Study Arm A (IFA):
Sterile saline solution 0.75ml w ill be added to 0.75 ml of LPV7 + tetanus peptide (total 1.5 ml). This w ill be combined w ith 1.5 ml of Montanide ISA-51VG to create an emulsion. Tw o (2) ml w ill be injected half-intradermally and half-subcutaneously in one skin location. Study Arm B (polyICLC): PolyICLC (Hiltonol) is provided from Oncovir, Inc. (Washington, D.C.) as a clinical grade reagent for experimental use in single-use vials containing 1 mL of a 2 mg/mL solution. We w ill administer 1 mg (0.5 ml) per vaccine. PolyICLC (Hiltonol) 0.75 ml w ill be added to 0.75 ml of LPV7 + tetanus peptide. One (1) ml of this w ill be injected halfintradermally and half-subcutaneously in one skin location. Study Arm C (resiquimod): Resiquimod is provided by 3M (Minneapolis, MN) and is packaged in multi-dose tubes (typically 3 g of 0.2% gel). Sterile saline solution 0.75ml w ill be added to 0.75 ml of LPV7 + tetanus peptide. One (1) ml of this w ill be injected halfintradermally and half-subcutaneously in one skin location. Immediately after vaccine injection, apply approximately 500 mg of 0.2% resiquimod gel w ith a gloved finger over a 25 cm 2 area of the skin at the vaccine site. Rub the gel into the skin for 1 minute until it vanishes. Follow ing application, the site may be covered w ith an occlusive dressing, such as Tegaderm, for about 8 hours. Participants are encouraged to w ash the area at 8 hours after application. Study Arm D (polyICLC + resiquimod): PolyICLC (Hiltonol) 0.75 ml w ill be added to 0.75 ml of LPV7 + tetanus peptide. One (1) ml of this w ill be injected half -intradermally and halfsubcutaneously in one skin location Immediately after vaccine injection, apply approximately 500 mg of 0.2% resiquimod gel w ith a gloved finger over a 25 cm 2 area of the skin at the vaccine site. Rub the gel into the skin for 1 minute until it vanishes. Follow ing application, the site may be covered w ith an occlusive dressing, such as Tegaderm, for about 8 hours. Participants are encouraged to w ash the area at 8 hours after application. Study Arm E (IFA + polyICLC): Part 1: PolyICLC (0.75 ml) w ill be added to 0.75 ml of LPV7 + tetanus peptide (total 1.5 ml). This w ill be combined w ith 1.5 ml of Montanide ISA-51VG to create an emulsion. Tw o (2) ml w ill be injected half-intradermally and halfsubcutaneously in one skin location.
Study Arm E2 (IFA + polyICLC): Part 2: PolyICLC (0.75 ml) w ill be added to 0.75 ml of LPV7 (total 1.5 ml). This w ill be combined w ith 1.5 ml of Montanide ISA -51VG to create an emulsion. Tw o (2) ml w ill be injected half-intradermally and half-subcutaneously in one skin location (as in study arm A).

Study Arm F (IFA + resiquimod):
Sterile saline solution 0.75ml w ill be added to 0.75 ml of LPV7 + tetanus peptide (total 1.5 ml). This w ill be combined w ith 1.5 ml of Montanide ISA-51VG to create an emulsion. Tw o (2) ml w ill be injected half-intradermally and halfsubcutaneously in one skin location (as in study arm A). Immediately after vaccine injection, apply approximately 500 mg of 0.2% resiquimod gel w ith a gloved finger over a 25 cm 2 area of the skin at the vaccine site. Rub the gel into the skin for 1 minute until it vanishes. Follow ing application, the site may be covered w ith an occlusive dressing, such as Tegaderm, for about 8 hours. Participants are encouraged to w ash the area at 8 hours after application. Study Arm G (IFA + polyICLC + resiquimod): PolyICLC (0.75 ml) w ill be added to 0.75 ml of LPV7 + tetanus peptide (total 1.5 ml). This w ill be combined w ith 1.5 ml of Montanide ISA-51VG to create an emulsion. Tw o (2) ml w ill be injected half-intradermally and halfsubcutaneously in one skin location (as in study arm A). Immediately after vaccine and polyICLC injection, apply approximately 500 mg of 0.2% resiquimod gel w ith a gloved finger over a 25 cm 2 area of the skin at the vaccine site. Rub the gel into the skin for 1 BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)

Storage of Prepared Peptide Vaccines
The prepared peptide vaccines w ill be stored in a plastic syringe and delivered to the clinicians in a plastic bag. This bag w ith the syringe w ill be stored at room temperature until the vaccine is administered. Ideally, the vaccine should be administered w ithin 1-2 hours after mixing. If the vaccine is not administered w ithin 4 hours after mixing, it should be discarded.

Drug Accountability and Distribution
Study drug w ill be accounted for and distributed to UVA patients and the outside sites using the HITC InvestMed Database. The InvestMed Database is an Access database that interfaces w ith the Cancer Center Clinical Trials clinical database so that drug may be allocated and dispensed on a per patient basis to UVA patients. For the outside sites, drug is dispensed to the corresponding institution and tracked on a per institution basis. Study drug w ill be sent to the outside sites by HITC personnel.

Designation of Vaccine Sites
Evidence suggests nodes proximal to a tumor site may be relatively immunosuppressed 64 ; therefore, the vaccination sites w ill be distant from sites of know n tumor w henever possible. In general, participants w ill be vaccinated in upper arm or thigh locations. If a participant has received a prior experimental vaccine, vaccines should not be administered at the same site as the prior experimental vaccine, but they may be administered on the same extremity as prior vaccines.

Regimen
Part 1: Each vaccine w ill be administered at a different site, rotating among available extremities w hen possible.
Part 2: The same skin location w ill be used for all vaccines. If the vaccine site has severe inflammation or ulceration after multiple vaccines, the next vaccine may be placed adjacent to the original site.
Chronic inflammatory reactions at the vaccine sites are expected to occur in all participants w ho receive Montanide ISA-51. Induration may persist for days to months, but is not expected to require additional therapy. Sterile abscesses may occur in some participants. These are not a basis for discontinuation of the vaccines. How ever, if the inflammatory reactions develop significant ulceration of the skin (> 2 cm), then this w ould be considered a dose-limiting toxicity and the patient w ould be taken off the study.

Vaccination doses:
Part 1: For each vaccine, 300 mcg of each of the long peptides in LPV7 and 200 mcg of the tetanus helper peptide w ill be administered, plus 1 ml IFA or the dose of TLR agonists as specified in Table_1A and section 5.2.1.
For each vaccine, 300 mcg of each of the long peptides in LPV7 w ill be administered, plus 1 ml IFA and the dose of TLR agonists as specified in Table 1B and Section 5.2.1.

Route of administration:
At each injection site for emulsions w ith IFA, a single needle puncture is to be performed, w ith delivery of 1 cc volume into the subcutaneous tissue and 1 cc volume into the dermis. This usually w ill require advancing the needle into the subcutaneous tissue to deliver the first 1 cc over an area w ithin reach of that needle, then pulling the needle back to near the skin puncture site, and advancing it into the intradermal tissue for delivery of the remaining 1 cc. For injections w ith peptides in polyICLC w ithout IFA, the total volume w ill be 1 ml, w hich w ill be divided betw een subcutaneous tissue and dermis as above.

Post-Vaccination Observation
All participants w ill be closely observed for adverse events for at least 20 minutes follow ing each vaccination. Any time thereafter, participants should report any adverse events to the research coordinator or research clinician. The biopsies w ill consist of three 4-mm punch biopsies of skin. Note that vaccine sites w ill be rotated to different skin sites for each vaccination; so there w ill not be repeat injection of vaccines into the prior skin biopsy sites.

Discontinuation
If participants are discontinued prior to one or more scheduled skin biopsies, they w ill not receive those biopsies. Protocol treatment w ill be discontinued for any of the follow ing reasons: 5.5.2.1 Any dose-limiting toxicity as defined in Section 7.4.

5.5.2.2
In circumstances w here assessment of an AE is limited, such as by intercurrent illness, or w hen laboratory studies are required to assess for other causes of toxicity, the vaccine/biopsy schedule may be interrupted for up to 7 days . Delay of one vaccine administration, biopsy, or evaluation visit by up to 7 days w ill not be considered a protocol violation, regardless of attribution. If more than one vaccine is delayed by 8 days or more, treatment must be discontinued.

5.5.2.3
Disease progression requiring other therapy (e.g. surgery under general anesthesia, radiation, chemotherapy, or steroid therapy). The appearance of small metastases or recurrent tumor deposits w ill not be a basis for discontinuing the vaccinations. Biopsy to determine the nature of new lesions, or minor surgical procedures to excise a new lesion, w ill not be a basis for discontinuing vaccinations.

Elective Withdraw al
A participant w ho is enrolled but neither receives any study drug nor has a skin biopsy on protocol may be replaced. Every attempt w ill be made to evaluate any data from these participants for endpoint assessment.

Delayed Visit for Reasons Other Than Toxicity
A schedule for return visits should be established at the first visit. If a participant misses a treatment, the missed treatment w ill be administered as soon as possible, so that the subsequent vaccinations are given in the appropriate intervals. Treatment may be continued for an additional time period, if needed. Participants w ho are vaccinated outside of the established schedule should return to the original schedule as soon as possible.
The table below defines w hat constitutes a delayed visit, w hether the participant should continue to be treated, and w hether a protocol violation/deviation should be reported and recorded. The range of days is counted from the original scheduled date.

Concomitant Medications
Medications taken in the month prior to randomization (Part 1) or arm assignment (Part 2) should be recorded on the baseline case report form. This includes prescription medications, over-the-counter medications, injected medications, biological products, blood products, imported drugs, or street drugs. Participants should be maintained on drugs that they w ere taking prior to entry unless a change in regimen is medically indicated.  FluMist) are permitted, but should be administered at least 2 w eeks prior to or at least 2 w eeks after a study vaccine. 5.6.2.7 Topical forms of chemotherapeutic reagents (e.g. Efudex ® ) are permitted, but should be administered at least 2 w eeks after a study vaccine.

Screening
The follow ing studies should be completed w ithin 6 w eeks prior to randomization unless otherw ise noted: 6.1.1 Class I HLA-typing (any time prior to randomization; typing for HLA -A and HLA-B is sufficient). If the patient is know n to express HLA-A1, A2, or A3, HLA-B testing is not required).
6.1.2 Review of pathology at the treating and participating institution (any time prior to randomization). Central pathology review at UVA is not required for pathology review ed at other participating institutions.

Tumor tissue collection
If during the study, participants develop metastases or recurrences, these may be removed, and follow ing receipt by pathology, may be evaluated by the study research team. Tissue samples may be screened for antigen expression or protein profiles using tes ts such as Western blots, immunohistochemistry, PCR, flow cytometry or gene chip analysis. Tumor escape mechanisms may also be evaluated. Specimens w ill be used in immunological assays to assess T cell function or antibody response. Assays generally used for this type of testing include, but are not limited to, ELIspot assays, ELISAs, chromium-release assays, proliferation assays and intracellular cytokine staining. Specimens may be used to study the immunologic aspects of the tumor microenvironment or as targets or controls in laboratory assays. Specimens may be used to establish cell lines for long-term studies.
This tissue may also be compared to lesions resected prior to enrollment, w hich w ill be requested from the pathology department of each institution as paraffin-embedded tissue samples, and these tissues may be banked for use in future studies. If participants are removed from the study or progress during or after follow -up, tissue may be collected for use as part of this study, as described above, or banked for use in future studies

REGULATORY AND REPORTING REQUIREM ENTS
Adverse event reporting requirements for this trial, as specified in this section, apply to patients accrued at all participating sites. There w ill be accrual at UVA and at MD Anderson Cancer Center; UVA w ill be the coordinating site, and responsibility for data coordination and reporting for the study as a w hole falls to the Human Immune Therapy Center (HITC). Throughout this section, w hen referring to 'Cancer Center Database', 'Cancer Center DSMC', 'CRC', 'PI' or, 'nurse practitioner', they are preceded by 'site' w hen it refers to reporting w ithin either of the participating institutions, or are preceded by 'UVA' w hen it refers to central reporting to the coordinating UVA center.
In addition to the reporting requirements for the study, specified in this section of the protocol, investigators should also follow their institutional policy for documentation in their database and reporting to the IRB locally at their ow n site.

Adverse event (AE) -
Any unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated w ith the use of a medical treatment or procedure regardless of w hether it is considered related to the medical treatment or procedure (attribution of unrelated, unlikely, possible, probable, or definite). Medical conditions or diseases present before starting the investigational drug w ill be considered as treatment-related AEs if they w orsen after starting study treatment.

Unexpected AE -
Any adverse event not listed in Section 7.4.3.

Serious AE -
Any adverse drug experience occurring at any dose that results in any of the follow ing outcomes:  death;  a life-threatening adverse drug experience;  inpatient hospitalization, or prolongation of existing hospitalization (as defined below in this section);  a persistent or significant disability/incapacity; or a congenital anomaly/birth defect.  Important medical events that may not result in death, be life threatening, or require hospitalization may be considered a serious adverse drug experience w hen, based upon medical judgment, they may jeopardize the patient or subject and may require medical or surgical intervention to prevent one of the outcomes listed in this definition.
Hospitalization for expedited AE reporting purposes is defined as an inpatient hospital stay equal to or greater than 24 hours. Hospitalization is used as an indicator of the seriousness of the adverse event and should be reserved for situations w here the adverse event truly fits this definition and not for hospitalizations associated w ith less serious events. For example, a hospital visit w here a patient is admitted for observation or minor treatment (e.g. hydration) and released in less than 24 hours w ould not be considered serious. Furthermore, hospitalization for pharmacokinetic sampling is not an AE, and therefore is not to be reported either as a routine AE or in an expedited report.

Unanticipated problem -
An unanticipated problem is any event/experience that meets ALL 3 criteria below :  Is unexpected in terms of nature, severity or frequency given the research procedures that are described in the protocol-related documents AND in the characteristics of the subject population being studied.  Is related or possibly related to participation in research. This means that there is a reasonable possibility that the incident may have been caused by the procedures involved in the research study. (see section 7.2)  The incident suggests that the research placed the subject or others at greater risk of harm than w as previously know n or recognized OR results in actual harm to the subject or others.

Protocol Violation-
A protocol violation is defined as any change, deviation, or departure from the study design or procedures of a research project that is NOT approved by the institution's IRB prior to its initiation or implementation, OR deviation from standard operating procedures, Good Clinical Practices (GCPs), federal, state or local regulations. Protocol violations may or may not be under the control of the study team or UVa staff. These protocol violations may be major or minor violations.
7.1.6 Suspected Adverse Reaction (as defined in 21 CFR 312.32 (a))-Any adverse event for w hich there is a reasonable possibility that the drug caused the adverse event.
7.2 Attribution Assessment 7.2.1 Attribution -The determination of w hether an adverse event is related to a medical treatment or procedure. The attribution groups are: Definite -Applies to those adverse events w hich, the investigator feels are incontrovertibly related to the vaccine. An adverse event may be assigned an attribution of definitely related if or w hen (must have all of the follow ing):  It follow s a reasonable temporal sequence from administration of the test drug.  It could not be reasonably explained by the know n characteristics of the subject's clinical state, environmental or toxic factors, or other modes of therapy administered to the subject.  It disappears or decreases on cessation or reduction in dose w ith re-exposure to drug. (Note: This is not to be constructed as requiring re-exposure of the subject; how ever, the group of definitely related can only be used w hen a recurrence is observed.)  It follow s a know n pattern of response to the test drug.
Probable -Applies to those adverse events for w hich, after careful consideration at the time they are evaluated, are felt w ith a high degree of certainty to be related to the test drug. An adverse event may be considered probably related if or w hen (must have three of the follow ing):  It follow s a reasonable temporal sequence from administration There are important exceptions w hen an adverse event does not disappear upon discontinuation of the drug, yet drugrelatedness clearly exists (e.g. bone marrow depression, fixed drug eruptions, tardive dyskinesia).  It follow s a know n pattern of response to the test drug.
Possible -Applies to those adverse events for w hich, after careful consideration at the time they are evaluated, a connection w ith the test drug administration appears unlikely but cannot be ruled out w ith certainty. An adverse event may be considered possibly related if or w hen (must have tw o of the follow ing):  It follow s a reasonable temporal sequence from administration of the test drug.  It could not readily have been produced by the subject's clinical state, environmental or toxic factors, or other modes of therapy administered to the subject.  It follow s a know n pattern of response to the test drug.
Unlikely -Applies to those adverse events for w hich, after careful consideration at the time they are evaluated, are judged to be unrelated to the test drug. An adverse event may be considered unlikely if or w hen (must have tw o of the follow ing):  It does not follow a reasonable temporal sequence from administration of the test drug.  It could readily have been produced by the subject's clinical state, environmental or toxic factors, or other modes of therapy administered to the subject.  It does not follow a know n pattern of response to the test drug.  It does not reappear or w orsen w hen the drug is readministered. Unrelated -Applies to those adverse events, w hich after careful consideration, are clearly and incontrovertibly due to extraneous causes (disease, environment, etc.).

Data collection
Data w ill be collected using a centralized electronic case report form called ON-line Clinical Oncology Research Environment = Oncore.

Adverse Event Descriptions and Grading Scales
The NCI Common Terminology Criteria for Adverse Events (CTCAE) version 4.03 w ill be used for the characterization and grading of adverse events (see Appendix 6).

Time Span for Reporting Adverse Events
Reporting of AEs w ill begin w hen the subject is administered the study drug or has a study related biopsy. Events occurring through 30 days after administration of the last vaccine, regardless of attribution, w ill be reported. AEs should be follow ed to resolution or stabilization. If an AE w orsens and becomes an SAE, it should be reported as serious per the guidelines specified for SAE reporting.
AEs that are possibly, probably, or definitely related to the vaccine w ill be recorded until the subject completes treatment follow -up. If, during treatment follow -up, the subject receives additional treatment, subjects w ill be off treatment follow -up and w ill be follow ed yearly for disease progression and survival.  Skin and subcutaneous disorders Other: any rash *NOTE: Hyperglycemia ≥ Grade 3 is not an expected event. If a non-fasting hyperglycemia adverse event occurs at a level ≥ Grade 3, a fasting serum glucose test will be ordered and the event will be re-graded appropriately. The fasting serum glucose measurement will be used for the purpose of determining reporting requirements.
Injection site reaction/induration w as experienced at maximum grade 3 in 5% of patients in Arm A of Mel44. Ulceration w as experienced as maximum grade 3 in 7% of patients in Arm A of Mel 44. By the rule above (observed in ≥5% of treated patients), injection site reaction, induration, and ulceration w ould be expected at grade 3. Despite this and in the interest of the patients, w e w ould consider grade 3 injection site reaction w ith significant ulceration of the skin (> 2 cm) as a DLT (see section 7.8).

Adverse Events Expected with Resiquimod application at the vaccine site.
The investigator brochure specifies the follow ing adverse events in more than 1800 patients treated w ith topical resiquimod in 25 clinical studies. Reported adverse events w ere primarily local adverse events including protocol-defined local skin reactions (including, erythema, edema, ulceration, vesicles, w eeping, scabbing, and flaking); and protocoldefined local symptoms at the site of administration (including pain, burning, pruritus, and numbness/tingling). Other adverse events at the application site include drainage, infection, and pigmentation changes. Systemic adverse events that may be related to resiquimod due to a systemic induction of cytokines and/or release of locally induced cytokines into the circulation include influenza-like symptoms (e.g., fever, headache, arthralgia, myalgia, chills, asthenia, fatigue), and transient decreases in neutrophils and total leukocytes that may be related to cellular trafficking to the application site. Other infrequent adverse events that w ere considered possibly or probably related to study medication included liver function test abnormalities, abdominal cramping, nausea, vomiting, diarrhea, backache, neck pain, rhinitis, urticaria, and dizziness.
The formal list, below , of expected AEs for resiquimod, for this study, is based on published data related to its clinical use 65;66 The table below is based on a study applying resiquimod 3 times per w eek for 4 w eeks, for treatment of actinic keratosis. This study had 4 treatment arms, using 0.01, 0.03, 0.06, and 0.1% resiquimod gel. We w ill be using 0.2% resiquimod gel, only once a w eek; so toxicities from 3x/w eekly administration w ith 0.06% resiquimod use may exceed w hat w e observe 65 . Overall in that study, no deaths or serious adverse events w ere reported, but some adverse events are described as severe. We expect grade 2 toxicities in these categories. Listed in the table below are those categories w here adverse events w ere recorded in 10% of patients or more, in any study group, but the percentages listed are those for patients receiving resiquimod 0.06% three times per w eek. Toxicities at sites of resiquimod application should be limited to grade 1, 2, or 3 (w ith ulceration ≤ 2 cm) dermatologic toxicities, and should be manageable either by continuation of resiquimod if the patient is tolerating the toxicities and if the toxicities are relatively stable, or they may be managed by holding vaccination for one to several days (up to 7 days w ithout being a DLT).

Adverse events expected with polyICLC (Hiltonol)
The follow ing expected toxicities for polyICLC are based on the data from 45 patients treated w ith 20 mcg/kg 3x/w eek. This w ould be about 1.4 mg per dose and 4.2 mg per w eek (much higher than w e w ill use).

Adverse events expected from vaccine and normal skin biopsies.
Below is a list of expected AEs related to skin biopsies:  The process of reporting AEs will occur as follows: A. Identify the type of event using the NCI CTCAE v4.03 (Appendix 6). The CTCAE provides descriptive terminology and a grading scale for each adverse event listed. B. Grade the event using the NCI CTCAE v4.03. C. Determine w hether the AE is related to the vaccine. Attribution groups are defined in Section 7.2.1. D. Determine the prior experience of the AE. Expected events are those that have been previously identified as resulting from administration of the vaccine. An AE is expected w hen it is listed in Section 7.4.3. E. Determine the adverse event classification of the AE, as specified and defined in Section 7.5. F. Determine w hether the AE is a dose-limiting toxicity and report as specified below . G. Determine w hether the AE is an SAE and report as specified below . H. AEs determined to require expedited reporting to the Sponsor, UVA HITC, must also be reported to your ow n institution according to the local policy and procedures.

Reporting Dose Limiting Toxicities All DLT's (defined in Section 7.8) should be reported by phone or fax to the HITC within 24 hours of when the site is notified of the event and entered into Oncore within 5 calendar days of when the site is notified of the event.
DLT's that are deemed serious and unexpected should be submitted to to a site's local IRB per their institutional guidelines.

Additional Reporting Requirements for the Sponsor (UVA)
Reporting to the FDA Serious and unexpected suspected adverse reactions w ill be reported to the FDA no later than 15 calendar days after the sponsor determines that the requirements for an IND safety report have been met. The FDA w ill be notified using an FDA Form 3500a. Unexpected fatal or life-threatening suspected adverse reactions w ill be reported to the FDA no later than 7 calendar days after the Sponsor receives the initial information of the event. The FDA w ill be notified using an FDA Form 3500a. Other adverse event information w ill be sent to the FDA in the IND annual report.

Reporting to the Sites
The sponsor w ill report serious and unexpected suspected adverse reactions to the outside sites no later than 15 calendar days after the sponsor determines that the requirements for an IND safety report have been met.

Reporting of Subject Withdraw als/Dropouts Prior to Study Completion
Subjects w ho w ithdraw consent and those dropping out of the study secondary to an AE w ill be reported to the UVA IRB yearly on the IRB continuation form. These data should be reported to the sponsor w ithin 30 days of subject w ithdraw al or dropout. 7.7 Adverse Event Review and Monitoring 7.7.1 In addition to clinic notes, adverse events w ill be initially captured using study-specific tools and participant toxicity diaries.
As specified in the study flow chart (Appendix 1), each participant w ill be evaluated by a licensed clinician. The follow ing w ill be performed as designated in the protocol: routine disease-directed physical exam including performance status, blood collection, reassessment for vitiligo, and examination of the skin and nodal basins for evidence of metastasis.
Participants w ill keep a daily diary of toxicities for days 1 through 85. The diaries w ill be review ed by a research clinician prior to the next scheduled vaccine. During clinic visits, participants w ill also be asked about subjective symptoms including headache, malaise, fatigue, dyspnea, nausea, rash, diarrhea, abdominal discomfort, peripheral nerve pain, visual changes, appetite, tremors, night sw eats, and ability to concentrate. Additional toxicities w ill be captured from laboratory tests. For each AE (w ith the exception of Grade 1 hematologic/metabolic events), date of onset, duration, grade, and attribution w ill be noted in the participant's study chart, on study documents, or in the clinic note, and w ill be entered into the UVA Cancer Center database.
After each vaccination, participants w ill be observed for AEs for at least 20 minutes. Followup phone calls w ill be made per the judgment of the research clinicians w ith regard to individual participant need. Subjects w ill be instructed on how to reach their provider should they have any questions and/or problems during the study.
In the event of an AE, appropriate action w ill be taken to ensure adequate care for the participant. If the participant is still on protocol, treatment delay or w ithdraw al from the protocol w ill be considered according to the protocol guidelines. 7.7.3 SAEs experienced by participants at all sites w ill be review ed tw ice a month by the sponsor during the UVA Melanoma Team meeting. This meeting w ill occur at least 20 times in a calendar year. Those present at the meeting include the sponsor/overall study PI, subinvestigators, protocol development staff, biostatisticians, research nurses, research coordinators, laboratory specialists, and laboratory research managers. These meetings may also include the review of individual participants to assess w hether they are protocol candidates, w hether AEs w arrant discontinuation, and w hether existing protocols should be continued or closed. 7.7.4 Monthly conference calls w ill be held w ith the Sponsor and the site designee to review protocol status, accruals, and adverse events. 7.7.5 The follow ing laboratory values w ill be recorded in the UVA Cancer Center database, graded using the CTCAE v4.03 (if a grading category exists), and reported as described in Section 7 Any abnormal laboratory values captured w hich are not included in the above list, but are considered to be pertinent positive clinical signs/symptoms, and laboratory results obtained as part of routine care of patients w ill be recorded in the UVA Cancer Center database and reported as described in Section 7.6. If there is any doubt on the part of study personnel concerning w hat constitutes a pertinent positive finding, the sponsor is to be consulted.

Adverse Event Stopping Guidelines
Based on results from previous vaccinations w ith 12-MP and tetanus peptide preparations, and w ith findings from use of resiquimod and polyICLC, significant toxicity is not anticipated. The study w ill be monitored continuously for treatment-related adverse events w ithin each vaccine preparation by the sponsor. Data from the UVA-Mel44 vaccine trial (Arm A) (using 12-MP and tetanus peptide) indicate an 8% rate of treatment-related unexpected Grade 3 adverse events. All adverse events Audit results Application of study designed stopping/decision rules Whether the study accrual pattern w arrants continuation/action Protocol violations 7.9.3 The UVA CC DSMC w ill meet every month for aggregate review of data. Tracking reports of the meetings are available to the PI for review . Issues of immediate concern by the DSMC are brought to the attention of the sponsor (and if appropriate to the PRC and IRB) and a formal response from the sponsor is requested. Per the UVA Cancer Center NIH approved institutional plan, this study w ill be audited approximately every 6 months.
7.10 Endpoint Data 7.10.1 Endpoint data w ill be collected using HITC IML data forms, participant-specific binders, and the HITC laboratory database.
7.10.2 The HITC laboratory database, w hich has passw ord-restricted access, is stored on the UVA Health System Computing Services secure server.

EVALUATION OF IMMUNOLOGIC RESULTS
A primary goal is to evaluate immunogenicity of LPV7 in each of 7 adjuvant preparations, including IFA, w ith or w ithout TLR agonists. The primary immunological response of interest is the CD8+ T cell response to the defined minimal peptides epitopes contained w ithin each of the 7 long peptides ( Table 2). This w ill be assessed primarily by ELIspot assay as outlined below . Associated analyses w ill include the durable immune response (w eek 26), and the peak magnitude of the CD8+ T cell responses. Details related to study design are in Section 9.
Studies in the peripheral blood w ill also assess CD4 T cell responses, T cell persistence, antibody response, Th1/Tc1 deviation and cytokine functionality. The antitumor immune response to vaccination w ith LPV7 w ith TLR agonists w ill be compared to vaccination w ith BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) IFA alone, and the highest immune response rate w ill be selected. Blood w ill be collected as described in Section 6.
Skin biopsy specimens from the vaccine sites after 1 (day 8) and 3 vaccines (day 22) (and from normal skin controls) w ill be evaluated for changes induced by TLR agonists, and these w ill be evaluated as w e have for a prior study 50 : immunohistochemistry of the vaccine-site immune cells to define the cell types and function; multiparameter flow cytometry to identify activation status of antigen specific cells, regulatory T cells, and MDSC; ELIspot assays for the ability to produce IFN; and Luminex studies to define the cytokine and chemokine profiles of the VSME.
The follow ing laboratory criteria w ill be used to evaluate effects of the vaccines and the TLR agonists in circulating peripheral blood mononuclear cells (PBMC) and in the vaccine sites.
In general, immunological evaluations w ill be performed in batch.

Immunohistochemistry and Immunofluorescence of skin biopsy specimens
For the immunohistochemical studies of the VSME, w e w ill employ stains w ith single antibodies (eg CD8), w ith multiple antibodies (eg CD8 and FoxP3), and a technique developed by our colleague Jim Mandell, w hich w e have w orking our lab now . This technique, called SIMPLE, enables repeated IHC stains on the same section, w ith w holeslide imaging using an Aperio or Leica system, and multiparameter evaluation of these cells in situ by combining the multiple images w ith each antibody. 67 Immunofluorescence studies w ill also be available in the event they are helpful.
We w ill obtain CD8/FoxP3 ratios to evaluate w hether higher (more favorable) ratios are induced by adjuvants that are associated w ith improved circulating immune responses.
For optimization of antibodies in IHC, w e have used positive control tissues, w hich w ill continue to be useful controls in the proposed studies. These include, but are not limited to, normal human peripheral blood lymphocytes (PBL), PBL activated w ith CD3/CD28 beads (activated PBL), human lymph node tissue from non-melanoma patients (LN), and inflamed human tissue from a surgical specimen of Crohn's disease (Crohn's).

ELIspot assay
We w ill evaluate CD4+ and CD8+ T cell responses in the peripheral blood using a direct (ex vivo) ELIspot assay for IFN as reported. 12 This w ill be performed in a core laboratory dedicated to immune monitoring, and w ith SOPs, controls, and quality assurance measures, w hich w e have reported. 12 . Vaccination site infiltrating lymphocytes w ill be evaluated by ELIspot as w ell. Vaccine site infiltrating lymphocytes w ill be compared to those in peripheral blood. ELIspot assays after one in vitro stimulation (stimulated ELIspot) may also be employed if appropriate, as reported 33 . CD4+ T cell responses w ill be assessed by IFN ELIspot assay using the mixture of 7 long peptides and also pools of 15-mer peptides (overlapping by 5 residues) from each peptide (4 peptides for each long peptide) as antigen sources.
Durability of CD8 and helper T cell responses w ill be assessed by measuring T cell activity at 26 w eeks, and for longer intervals w hen available, w ith the hypothesis that there w ill be increased durability of T cell responses w hen compared to prior experience w ith short peptides 13 and w ill be improved w ith TLR agonists w hen compared to IFA alone in this study. We have extensive stores of cryopreserved T cells reactive to minimal epitope peptides included in the 12 short peptides (12MP) mixture, and detectable by IFN-gamma ELIspot assay 8;12 and by cytotoxicity assays. 68;69 . These cells w ill serve as positive controls for reactivity to the minimal epitopes contained w ithin the long peptides used in this trial.
8.3 Tetramer assay and multiparameter flow cytometry Peripheral blood lymphocytes and VSME lymphocytes w ill be evaluated by flow cytometry after incubation w ith MHC-peptide tetramers or pentamers for the number of peptidereactive T-cells. We have validated tetramers for all 5 of the minimal CD8 epitopes restricted by HLA-A2 or A3 among the LPV7 peptides. We w ill use these MHC multimers w ith w hich w e also have substantial experience. 12 Tetramers w ill likely detect more antigen-specific T cells than are detected in the functional ELIspot assay, but the primary measure of CD8 T cell response w ill be the ELIspot assay, on w hich the statistical pow er w as based. Tetramers w ill also be used in multiparameter flow cytometry to determine the effector/memory status of the antigen-specific T cells 70 (CD45RA, CD27, CD28, CCR7, CD62L), and evaluation for multiple functions of the CD8 T cells (degranulation (CD107a), IFN, MIP-1β, TNF, and IL-2) 71 We w ill also use multiparameter flow cytometry to measure myeloid suppressor cells (CD14+ HLA-DR lo/neg), and several subtypes of MDSC in PBMC to determine if they are increased or decreased by any of these vaccine approaches, compared to our current experience w ith MDSC numbers using our short 12MP vaccine (data not show n).

Proliferation assay / Cytokine Analysis
To characterize responding T cells more completely, w e w ill evaluate w ith multiparameter flow cytometry after short-term culture w ith antigen. CD4 + T cells responding to antigen by CFSE dilution w ill be assayed for intracellular cytokine production to evaluate Th1/Th2/Th17 bias (Th1: IL-2, IFN, TNF; Th2: IL4, IL-5, IL-10; Th17: IL-17a). Antigen specific T-regs w ill also be assayed as described 72 .
We w ill also assay supernatants of PBMC, and of immune cells infiltrating the VSME, cultured 2 and 5 days w ith the 7 long peptides to define secreted cytokines and chemokines representative of the w hole cell population. We anticipate that the long peptides w ill induce Tc1 and Th1 dominant responses, w ith a high proportion of memory phenotype after several vaccines.
Peripheral blood mononuclear cells (PBMC) and vaccine-site inflammatory cells from participants in this clinical trial may also be evaluated for CD4 T cell reactivity, using a 3Hthymidine uptake assay, CFSE dilution or other assays as summarized above.
Responses to the tetanus helper peptide w ill be assayed as w ell as responses to the 30-mer melanoma peptides and 15-mer pools comprising those peptides. Either or both of tw o peptides w ill be used as negative controls for CD4 + T cell responses: one being the PADRE peptide (aK(X)VAAWTLKAa), w here X = L-cyclohexylalanine 73;74 .

Antibody responses
Antibody responses to 30 mer peptides (and to pools of 15-mer peptides) w ill be measured by ELISA assay using methods described by Sabbatini et al 2 .

Evaluation of tumor
Tumor tissue collected prior to enrollment or at the time of progression w ill be evaluated by routine histology and immunohistochemistry. In addition, in vitro evaluations of tumor tissue and tumor infiltrating lymphocytes may be completed.

STATISTICAL CONSIDERATIONS
This is an open-label, Phase I/II study evaluating the safety and immunogenicity of a long peptide vaccine (LPV7) + TLR for resected stage IIB -IV melanoma. There are tw o parts to the study.
For Part 1, the trial w as designed to find the range of optimal combination therapies, defined as combinations w ith high immunologic response and an acceptable level of toxicity. The primary outcomes included the frequency of treatment related dose limiting toxicities (DLTs) and frequency of immunologic response. For Part 2, the goal is to obtain preliminary safety and immunogenicity data on the optimal combination identified in Part 1 w hen administered in the same skin site for all 6 vaccines (same site vaccination).
For part 2, the vaccine regimen is selected based on available data from Part 1, in w hich the strongest immune responses w ere observed in participants enrolled on Arms E and G (Table 1A). Resiquimod is no longer available; so the regimen for Arm G is not currently feasible. Also, the total enrollment on Arm E is larger (15 vs 6) so that outcomes from Part 2 w ill be evaluable in a similar sample size as for Part 1, Arm E. This group w ill be called Study Arm E2.

Study design Part 1 (completed)
In order to assess the first tw o objectives, the co-primary objective is to determine a range of optimal combination therapies, among the combinations provided in the follow ing table w here an optimal combination is a combination that is estimated to have an acceptable toxicity profile as measured by DLT's and a high rate of immunologic response (IRR) as measured by peak immune response. DLTs are defined in Section 7.8 of the protocol.
Within the range of optimal doses a co-primary objective is if more than one combination is contained w ithin the range of optimal combinations, to estimate the difference in immunologic response rates betw een combinations, among those in the range of optimal combinations.
BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)

Combination allocation Part 1 (completed)
The goal is to determine the range of optimal combinations, w here optimal combination allocation w ill occur in tw o stages. The initial stage w ill accrue eligible participants in cohorts of one on each arm per the allocation process in 9.2.1 until a participant experiences a DLT. The second stage w ill allocate eligible participants based upon a continual reassessment method (CRM) 81 for combinations of agents 82 . Randomization w ill be based on equal allocation among allow able arms unless a w eighted allocation scheme is triggered (Section 9.2.1). Randomization w ill not be stratified by institution. Participants must be observed for a minimum of 3 w eeks after the initial vaccine for initial escalation betw een zones, and betw een the 1 st and 2 nd participants w ithin an arm. has not yet been tried. d) Escalation to a Zone 3 occurs only w hen all combinations in Zone 2 have accrued at least 1 participant and no participants have experienced a DLT.
 Allocation in Zone 3: a) Note Zone 3 contains a single combination, so the first eligible participant on Zone 3 w ill be entered onto combination G. b) If 0/1 participants experience a DLT, the next eligible participant w ill be entered onto combination G. If 0/2 participants experience DLT, the next eligible participant w ill be randomized to any combination. c) In the absence of DLT's, participants w ill continue to be randomized to combination A through G until a DLT occurs (upon w hich the modeling stage begins), or until at least 3 participants have been treated at every combination, at w hich point allocation w ill be based on w hich combination has the highest IRR. In the absence of DLT's, participants w ill continue to be entered on the combination w ith the highest IRR until sufficient information has been obtained regarding the BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) optimal combination, according to the stopping rules described below . Once a DLT has been observed, Stage 2 using CRM modeling begins.

Combination allocation in Stage 2:
The second stage w ill allocate eligible participants based upon a continual reassessment method for combinations of agents (R package, pocrm). The modeling stage uses (a) a selected set of possible orderings for the DLT probabilities and (b) a w orking model for the DLT probabilities under each ordering.  Within each ordering, the continual reassessment method (CRM) is fit, using the w orking model and the accumulated data. The second stage w ill accrue eligible participants in cohorts of one and use a CRM model fit to estimate DLT probabilities at each arm combination. For each order, = 1, ⋯ , 6, in Table 16, the DLT probabilities for each arm combination = 1, ⋯ , 7, are modeled via a one-parameter pow er model, (DLT at combination ) ≈ , w here the are the w orking model values for order given in Table 16. After accrual of each participant into the trial, the parameter is estimated for each ordering by maximum likelihood estimation w here the likelihood is given w here = the number of DLTs and =the number of treated participants in arm combination . The order w ith the largest likelihood is chosen and, w ithin this ordering, DLT probability estimates are updated for each combination. If there is a tie betw een the likelihood values of tw o or more orderings, then the selected order is randomly chosen from among the tied orderings. These DLT probabilities w ill be used to define a set of "acceptable" combinations, defined as any dose w ith estimated DLT probability less than or equal to 33%. This toxicity tolerance of 33% w as chosen based on the expectedness of adverse events.

Combination recommendation:
Once the set of "acceptable" combinations is determined, the recommended combination w ill be based upon how many participants have been entered into the study to that point. If less than 3 participants have been treated on any acceptable dose, then the recommended combination for the next entered participant w ill be chosen at random from the acceptable combinations. If at least 3 participants have been treated at every acceptable combination, the recommended combination for the next participant entered is defined as the "acceptable" combination w ith the highest observed IRR. After each participant, a new recommended combination is obtained, and the next entered participant is allocated to the recommended combination. Part 1 of the trial w ill stop once sufficient information about the optimal dose range has been obtained, according to the stopping rules outlined in Section 9.6.1.

Statistical properties Part 1 (completed)
Simulation results w ere run (R package, pocrm) to display the performance of the design characteristics. For each scenario, 1000 simulated trials w ere run. Each table reports the true DLT probability at each combination, the true IRR at each combination, the percentage of trials in w hich each combination w as recommended as the optimal combination, and the average number of participants treated in a simulated study. 9.3.1 Scenario 1: All true DLT probabilities are safe (i.e. less toxic than 33%) and one combination has the highest IRR. The true optimal combination is indicated in italicized bold type.

Scenario 2:
One combination (G) has true DLT probability more toxic than 33% and the three combinations in Zone 2 have the highest IRR among safe combinations. The true range of optimal combinations is indicated in italicized bold type.   Maximum sample size is based upon acquiring sufficient information to assess the objective of selecting the arm w ith the highest IRR, assuming at least one optimal combination has been found.
Data from our prior Mel43 trial in the arms not getting GM-CSF 12 indicated a high T cell response (5-fold increase in reactivity, at least 1200 cells per 100,000 CD8, and no overlap  63%)). Using the range as an approximate guideline, maximum sample size is established to ensure that if IRR among all seven arms differ by 30%, w ith the smallest expected rate of 50%, then the arm w ith the higher IRR w ould be selected w ith high probability, P=0.90. The initial maximum target accrual of 20 eligible participants per arm is based upon the ranking and selection procedure, BSH 75 , w hich allow s selection of the best treatment w orthy of further investigation in single-factor Bernoulli response experiments. How ever, it is not anticipated that all arms w ill fall in the range of optimal combinations, thus, sample size is estimated from the simulations and w ill be determined by the stopping rules below . Although maximum accrual is set at 140, in the simulation results below , on average a total of betw een 7 and 46 participants w ere required to complete the study.
Accrual is estimated at 2 participants per month at each site (4 per month overall). The maximum accrual to the study is estimated from the simulations to be approximately 52 participants. Adjusting for 10% ineligibility/dropout/w ithdraw al rate, maximum total target accrual for Part 1 is estimated at 58 participants. 9.5.2 Part 2 Immune response w as observed in 6/16 (37.5%, 90%CI [17.8, 60.9%]) from Part 1. For Part 2 target accrual is set at 16 eligible participants. At study conclusion, if w e observe that at least 12/16 (a doubling of the response rate) eligible participants satisfy the definition of immune response then the low er limit of a one-sided 80%CI around the observed immune response rate for Arm E2 w ould be 61.5% w hich w e w ould conclude supports that the immune response rate is likely to be higher for single site injections. If few er responses are observed then additional studies w ill be needed to better understand the impact of site rotation w ith the use of the tetanus helper peptide.
Accrual is estimated at 2 participants per month. Adjusting for 10% dropout rate, maximum total target accrual for Part 2 is estimated at 18 participants. 9.6 Stopping rules: 9.6.1 Part 1 (completed) The study w ill be stopped according to the follow ing rules: 1. The study w ill stop early for safety if the first three entered participants in Zone 1 experience DLT on combinations in Zone 1 in Stage 1. 2. If at any point in Stage 2, the set of acceptable combinations is empty, the trial w ill stop for safety. 3. The study w ill stop if the recommendation is to assign the next participant to a combination that already has 20 participants treated at that combination.  All participants w ho receive any study intervention w ill be monitored for adverse events. Adverse events w ill be described and coded based upon the NCI CTCAE v4.03. A DLT is defined as any unexpected adverse event that is possibly, probably or definitely related to treatment and satisfies the criteria in section 7.8. Occurrence of DLTs w ill guide escalation and stopping decisions. At study conclusion frequency, proportion and severity of adverse events, and DLTs by arm w ill be tabulated. 9.7.2 Immunologic response: Subjects are evaluable for immune response if at least one post treatment sample is measurable for response. Peak immunogenicity w ill be evaluated by direct ELIspot assay against the defined nonamer peptides in the peripheral blood. The presence or absence of an immune response w ill be defined by induction of at least a 2-fold increase in IFN-gammasecreting cells over background, and over pre-existing responses, and at least 30 IFNgamma secreting spots per 100,000 CD8 T cells, as described 34 . This assessment w ill be performed on blood through w eek 12, although a minimum of 4 w eeks of data w ill be used to guide decisions about the range of optimal dose combinations. IRR is defined as the proportion of participants w here presence (as defined above) of an immune response has been observed.
Durable immunogenicity w ill be evaluated w eek 26 by ELIspot assay against the defined nonamer peptides in the peripheral blood. The presence or absence of an immune response at that time w ill be defined by induction of at least a 2-fold increase in IFN-gamma-secreting cells over background, and over pre-existing responses, and at least 30 IFN-gamma secreting spots per 100,000 CD8 T cells, as described 34 . This assessment w ill be performed on blood through w eek 26.

At study conclusion:
Frequency and magnitude of adverse events w ill be summarized by arm. Point estimates and confidence intervals w ill be calculated for all dichotomous endpoints. If more than one combination is contained w ithin the range of optimal combinations, estimates and confidence intervals w ill be used to estimate the difference in IRR betw een pairs of combinations among those in the range of optimal combinations. Immunogenicity w ill be evaluated by direct ELIspot assay against the defined nonamer peptides in vaccine site biopsies using the above criteria. Disease-free survival is defined as the duration of time from start of treatment to time of recurrence/progression or death from any cause, w hichever occurs first. Participants w ho do not experience an event (recurrence/progression or death) w ill be censored at date of last contact. Overall survival is defined as the duration of time from start of treatment to time of death from any cause. Participants w ho do not experience an event (death) w ill be censored at date of last contact. Patients with cardiac disease resulting in marked limitation of physical activity. They are comfortable at rest. Less than ordinary activity causes fatigue, palpitation, dyspnea, or anginal pain.
Objective evidence of moderately severe cardiovascular disease.

Class IV.
Patients with cardiac disease resulting in inability to carry on any physical activity without discomfort. Symptoms of heart failure or the anginal syndrome may be present even at rest. If any physical activity is undertaken, discomfort is increased.
Objective evidence of severe cardiovascular disease.

Appendix 5: Lot Testing A. Preparation of the synthetic melanoma and tetanus peptides
All peptides have been synthesized under GMP conditions by Polypeptide Group (San Diego, CA) or comparable provider. Certificates of analysis and technical summaries for each of the peptides w ill be included in the chemistry and manufacturing portion of the IND application.
Peptide preparation and vialing has been performed under GMP conditions by Clinalfa/Merck for the tetanus peptide, and Bachem for the NY-ESO-1 long peptide, and w ill be performed under GLP conditions by the UVA Human Immune Therapy Center laboratory for the 6 peptides in the LPV6 long peptide mixture. Documentation relating to the procedures used to prepare and vial the peptides w ill comply w ith the CFR and w ill be included in the Chemistry and Manufacturing Section of the IND application.
B. Lot Release Quality Assurance Testing Prepared peptides w ill be subjected to the follow ing tests: 1. Identity: Identity w ill be confirmed by structural studies. The individual peptides w ill be tested for identity by mass spectrometry (to define molecular mass and amino acid sequence) and high performance liquid chromatography (HPLC, to confirm purity) in a GMP laboratory (Polypeptide Group). Amino acid analysis w ill also be performed to confirm identity further.
Before mixing, the amino acid sequence of each individual peptide preparation w ill be reconfirmed by mass spectrometry or co-elution.
After combining the peptides, the mixture w ill be subjected to HPLC analysis.
2. Purity: Purity w ill be assessed before and after vialing the peptide mixtures. Before vialing the peptide mixtures, each synthetic peptide w ill be evaluated for the presence of a single dominant species by high performance liquid chromatography (HPLC). Purity of each peptide component w ill exceed 90% (expected 94%-98%).
Variants of the original peptide may include incomplete products of synthesis, minor degradation products due to oxidation of methionine residues, and dimerization of cysteine-containing peptides. After vialing the peptide mixture, purity w ill be reconfirmed by HPLC.
3. Aggregation: Aggregation of peptides w ill be assessed by HPLC. Aggregation is not expected.

Trifluoroacetic acid (TFA):
The amount of total fluorine in each peptide preparation w ill be less than 0.5% or 5000 ppm. C. Ongoing Quality Assurance Testing Stability: We w ill follow guidelines for quality assurance studies to assure stability of the peptide preparation over time during the trial 76 .
The peptide preparations w ill be assayed for stability every six months for 2 years and every year thereafter. The follow ing analyses w ill be performed to confirm stability: 1. HPLC: HPLC w ill be performed to confirm purity. A comparison to previous HPLC data w ill be performed. Ideally, the purity of each peptide component w ill exceed 90% (94%-98%). Variants of the original peptide may include incomplete products of synthesis, minor degradation products due to oxidation of methionine residues, and dimerization of cysteine-containing peptides. Such minor variants w ill be tolerated as long as the intended peptide represents at least 75% of the total peptide species. With liquid formulations of this peptide preparation, w e have observed dimerization of cysteine-containing peptides, yet this peptide retained immunogenicity even w hen as little as 10% remains as monomer. For the present study, dimerization of cysteine-containing peptides w ill be acceptable as long as at least 25% remains as monomer by HPLC. Because measures of peptide quantity are subject to variability, a peptide lot w ill be rejected only if tw o sequential measures fail to meet the criterion stated above. If a second measure is required, the lot w ill be held until the second reading is obtained and show n to be acceptable.
2. Aggregation. Aggregation w ill also be assessed by HPLC.
3. Mass spectrometry. This w ill be performed on each peptide peak to define the molecular sequence and mass if there is any ambiguity about the identity of the peptides based on HPLC. New peaks on HPLC, or significant change in the elution profiles w ill w arrant determination of the peptide identities by mass spectrometry on each peak. 2) Section 1.1: updated to address current treatments for resected melanoma in the adjuvant setting. 3) Section 1.2.6: updated to address ipilimumab approval for treatment of resected melanoma in the adjuvant setting. 4) Section 4.1: removed reference to interferon packet. 5) Section 5.6.1: added targeted molecular therapy (e.g. vemurafenib, other inhibitor of mutant BRAF, MEK, or cKit) as nonmerpitted treatments. 6) Section 5.6.2.6: edited to indicate that any live vaccines (e.g. FluMist) are permitted, but should be administered at least 2 weeks prior to or at least 2 weeks after a study vaccine. 7) Updated reference list. 06-26-17 1) Updated cover sheet 2) Updated study personnel 3) Updated formatting and numbering throughout study document 4) Additional administrative changes and typographical errors corrected throughout the document to improve readability. 5) Table of Contents: Updated 6) Protocol precis  Design: The study has been divided into Part 1 and Part 2. Part 1 required rotating vaccine sites and accrual to this part was completed. Part 2 of the study has been added and the regimen requires that each vaccine is administered in the same skin site for all 6 vaccines.  The objectives have been revised to add objectives specific for part 2.  Table 1 has been re-labeled as Table 1A. The title of the table has been revised to specify that Table 1A is relevant for Part 1 of the study.  Addition of Table 1B: Study Arm E2 has been specified for Part 2 of the study.  The rationale for choosing the vaccine regimen from study arm E has been added.  Regimen: Revised to describe regimen for Parts 1 and 2 of the study.  Vaccine site biopsies: revised to specify that 3 punch biopsies of normal skin will be completed in the first 6 patients enrolled in Part 1 of the study.  Leadership: Revised to specify that Part 2 of the study will be conducted just at the University of Virgina.  Accrual: Adjusted to allow for an additional patients to be enrolled to part 2 of the study. 7) Section 1.0: Updated accrual information. 8) Section 1.1: Updated information describing incidence of melanoma and deaths from melanoma in 2017 and added a section on vaccine site selection to provide a rationale for evaluating same site vaccinations. 9) Section 1.2.6: Updated information on approved adjuvant therapies for melanoma. 10) Section 1.2.10: Added prior experience with same site vaccinations. 11) Added Table 5 to describe vaccine site reactions for prior Mel44 and Mel58 studies. 12) Section 2.0: Revised objectives to add objectives for Part 2.
BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) J Immunother Cancer doi: 10.1136/jitc-2021-003220 :e003220. 9 2021; J Immunother Cancer , et al. Patel SP 13) Section 3.1: Regimen-added the regimen for Part 2, which includes same site vaccinations. 14) Section 4.2.7(1): Corrected section reference. 15) Section 4.4.2: Revised randomization section and specified that Part 2 of the study will not include randomization. 16) Section 5.2.1: Added a description for preparing drug for Study Arm E2. 17) Section 5.4.2: Deleted prior text describing the vaccine regimen and replaced with a description for the administration of vaccines for Part 1 and Part 2 and a description of expected vaccine site toxicities. 18) Section 5.4.3: A section describing doses for part 2 of the study has been added. 19) Section 5.4.6 , 6.2.2, 6.2.4, Appendix 1: Clarified that biopsies of normal skin will be completed in the first 6 participants who are enrolled in Part 1. 20) Section 5.6: Clarified that patients will be randomized in Part 1 and assigned to an arm in Part 2 of the study. 21) Section 7.4.3: Clarified that grade 3 injection site reactions with significant ulceration of the skin (> 2 cm) will be considered DLTs. Clarified dermatologic toxicities for resiquimod. 22) Section 7.8: Added grade 3 injection site reaction with ulceration ≤ 2 cm as an exception to the DLT rule 23) Section 9.0: Revised introduction to statistical considerations to add information about Part 2 of the study. 24) Section 9.1-removed the DLT rule from this section and revised to refer to section 7.8 of the protocol. 25) Sections 9. 2, 9.3, 9.4, 9.5, 9.6: clarified which sections pertain to Part 1 and added statistical language for Part 2. The dosing for resiquimod was revised to include an acceptable dose range of 400-600 mg. Instructions about coverage and washing the affected area have been incorporated into the study protocol. 4) Section 1.2.8: Reference added 5) Table 3: Reformatted 6) Table 4: Updated with new data 7) Section 4.1.4: Clarification of inclusion criterion and whether repeat scans are needed after stereotactic radiotherapy. 8) Section 4.1.5: Clarification of inclusion criterion and the description of required lymph node basins. 9) Section 4.1.8: Clarification of inclusion criterion a) specifications for patients with Gilbert's disease and b) re-defined acceptable limit for Hgb-A1C level. 10) Section 4.2.11: Addition of the following to allowable autoimmune disorders: Hypothyroidism of any etiology on stable thyroid hormone replacement therapy. 11) Section 5.2.1: Arms E and G: Change in mixing instructions for PolyICLC.
PolyICLC will be incorporated into the vaccine mixture rather than administered as a separate injection. 12) Section 5. 2) The Table of Contents was updated. 3) The dose of resiquimod has been corrected throughout the study document. The correct dose is 1000 mcg and the drug will be supplied in multi-dose tubes. Section 1.2.6 also includes information on prior usage with this dose. 4) Section 4.1.4: An editorial change was made and also a clarification that the brain metastases in this section refer to treated brain metastases. 5) Section 4. Information about the interferon education packets was changed to include a statement that participants mus t convice the study team that the answered the questions accurately or understand the correct answers. 8) Section 5.4.1: A clarification was made to this section to specify that chronic inflammation is expected in subjects who receive Montanide ISA -51. Also, editorial changes were made to this section. 9) Section 6.2.2: "Vaccine site biopsy (3)" was removed from the Day 8 schedule of events as the biopsy procedures were listed twice for this day. 10) Sections 6.2.1, 6.2.2, 6.2.3, 6.2.4, 6.2.5, 6.2.6, 6.2.7, 6.2.8: Clarfication was made to the language pertaining to the interval history and physical exam. Examination of vaccine sites was added to each of the study visit days, when applicable. 11) Section 7: Editorial correction was made to change "or" to "for". 12) Section 7.4.3: A redundant section related to AEs that do not require expedited reporting. Reporting guidelines are provided in Table 12. 13) Section 7.4.3: A clarification was made to the dosing regimen that was used for resiquimod in prior studies. 14) Section 7.6.1: A clarification was made to the process of reporting AEs. AEs determined to require expedited reporting to the sponsor, must also be reported to a site's institution according to local policy and procedures.
15) Section 7.6.3: A clarification was made regarding reporting AEs to the sites. Serious and unexpected suspected adverse reactions will be reported to the sites no later than 15 calendar days after the Sponsor determines that the the requirements for an IND safety report have been met. 16) Section 7.7.2: The phrase "and at UVA by the UVA melanoma team" was removed from this section as it did not apply to the outside sites. Individual AEs will be reviewed at each site. 17) Section 7.7.3: The term "weekly" was removed when describing the melanoma team meetings. The melanoma team meets 20 times a year, as specified in section 7.7.4 of the DSMP. The term "may" was included in the description of the events that are reviewed at each meeting as the agenda varies depending upon study accruals. 18) Section 7.7.4 : The DSMP was modified to specify that month conference calls will be held between the Sponsor and the sites. 19) Section 9.2.2: "R-package, pocrm" was added for clarification. Additional statistical methods were added to clarify the model of DLT probabilities. 23. Appendix 1: Administrative changes have been made to the X-page to improve readability. 24. Additional administrative changes and typographical errors corrected throughout the document to improve readability. 08-06- 13 1) An additional study group has been added in response to the recommendations of the FDA and of our collaborators at MD Anderson Cancer Cente 2) The study design has been changed to incorporate an adaptive design in order to minimize data with the least number of patients. This reduces the target sample size by half.
3) Section 4.4 of the study protocol has been modified to include additional details regarding the registration and randomization of study patients. 4) A section for study drug accountability and distribution has been added (section 5.3) 5) Table of Contents has been updated.
6) The source of Hiltonol (polyICLC) has been updated, in section 1.2.5. 7) Updated section 1.2.10 to specify that the peptides for the vaccines have been synthesized under GMP conditions, rather than GLP. 8) Appendices about vaccine preparation, immunologic assays, and details of tissue handling with the biopsies has been removed from the protocol and will be included ina Study Manual and Investigator Brochure. 9) Numerous small administrative changes have been made to improve clarity, consistency and readability and to provide appropriate citations. 10) Details about the quality assurance studies to be done on the peptides have been added to respond to the FDA after our preIND meeting. 11) Skin biopsies have been amended: Patients 1-6 will have 3 skin punch biopsies rather than 2, on day 1. The 3rd sample from skin on day 1 in those patients and from all patients on days 8 and 22 will be stored in RNA later rather than be collected for viable cell studies. 12) Signature page and study staff contact information has been added. 13) X-page has been corrected with regards to the timing of the normal skin biopsies.
12/05/11 1) Section 4.4 of the study protocol has been modified to include additional details regarding the registration and randomization of study patients.
2) A section for study drug accountability and distribution has been added (section 5.3) 3)