Article Text
PDF
Abstract
Background In patients with stage III melanoma, despite surgical resection and adjuvant systemic therapy, locoregional recurrences still occur. The randomized, phase III Trans-Tasman Radiation Oncology Group (TROG) 02.01 trial demonstrated that adjuvant radiotherapy (RT) after complete lymphadenectomy (CLND) halves the incidence of melanoma recurrence within local nodal basins without improving overall survival or quality of life. However, the study was conducted prior to the current era of adjuvant systemic therapies and when CLND was the standard approach for microscopic nodal disease. As such, there is currently no data on the role of adjuvant RT in patients with melanoma who recur during or after adjuvant immunotherapy, including those that may or may not have undergone prior CLND. In this study, we aimed to answer this question.
Methods Patients with resected stage III melanoma who received adjuvant anti-programmed cell death protein-1 (PD-1) (±ipilimumab) immunotherapy with a subsequent locoregional (lymph node and/or in-transit metastases) recurrence were retrospectively identified. Multivariable logistic and Cox regression analyses were conducted. Primary outcome was rate of subsequent locoregional recurrence; secondary outcomes were locoregional recurrence-free survival (lr-RFS2) and overall RFS (RFS2) to second recurrence.
Results In total, 71 patients were identified: 42 (59%) men, 30 (42%) BRAF V600E mutant, 43 (61%) stage IIIC at diagnosis. Median time to first recurrence was 7 months (1–44), 24 (34%) received adjuvant RT and 47 (66%) did not. Thirty-three patients (46%) developed a second recurrence at a median of 5 months (1–22). The rate of locoregional relapse at second recurrence was lower in those who received adjuvant RT (8%, 2/24) compared with those who did not (36%, 17/47, p=0.01). Adjuvant RT at first recurrence was associated with an improved lr-RFS2 (HR 0.16, p=0.015), with a trend towards an improved RFS2 (HR 0.54, p=0.072) and no effect on risk of distant recurrence or overall survival.
Conclusion This is the first study to investigate the role of adjuvant RT in patients with melanoma with locoregional disease recurrence during or after adjuvant anti-PD-1-based immunotherapy. Adjuvant RT was associated with improved lr-RFS2, but not risk of distant recurrence, demonstrating a likely benefit in locoregional disease control in the modern era. Further prospective studies are required to validate these results.
- melanoma
- adjuvants, immunologic
- immunotherapy
- radiotherapy
- immune checkpoint inhibitors
Data availability statement
Data are available upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
There is an absence of prospective clinical trial data on the efficacy of adjuvant radiotherapy (RT) for patients with melanoma who relapse during or after adjuvant immunotherapy, as the landmark Trans-Tasman Radiation Oncology Group (TROG) 02.01 trial was conducted prior to the current era of adjuvant systemic therapies.
WHAT THIS STUDY ADDS
This is the first study to investigate the role of adjuvant RT in patients with melanoma with locoregional disease recurrence during or after adjuvant anti-programmed cell death protein 1-based immunotherapy. Results demonstrate that adjuvant RT provides a benefit for locoregional disease control in the modern era of adjuvant systemic therapy.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
This study provides important preliminary data to support the use of adjuvant RT for locoregional disease control in appropriately selected patients with disease recurrence during or after adjuvant immunotherapy, and highlights the need for further prospective studies to validate these results.
Background
In patients with stage III melanoma, despite surgical resection and adjuvant therapy, locoregional recurrences can still occur. Significant morbidity can result from such recurrences, including pain, ulceration, discharge, lymphedema and reduced limb function. These risks are increased in those with a high number of involved lymph nodes (LNs), large deposits of tumor within the LN and those with extranodal extension (ENE). While adjuvant systemic therapies have significantly improved patient outcomes in recent years, approximately 50% of patients still develop disease recurrence despite this.1–3
The randomized, phase III Australia and New Zealand Melanoma Trials Group (ANZMTG) 01.02/ Trans-Tasman Radiation Oncology Group (TROG) 02.01 trial investigated whether adjuvant radiotherapy (RT) after complete cervical, inguinal or axillary lymphadenectomy reduced nodal field recurrence in patients with high-risk macroscopic stage III disease.4 After a median follow-up of 73 months, 21% of patients who received adjuvant RT developed nodal field recurrence, compared with 36% of patients who underwent observation only (HR 0.52, 95% CI 0.31 to 0.88, p=0.023). The trial therefore demonstrated that adjuvant RT essentially halves local LN involvement at first recurrence.4 Of note, recurrence-free survival (RFS) and overall survival (OS) were not statistically different between the adjuvant RT and observation groups: HR 0.89 (95% CI 0.65 to 1.22, p=0.51) and HR 1.27 (95% CI 0.89 to 1.79, p=0.21), respectively. Furthermore, there was no significant difference in quality-of-life measures between the two treatment groups.4 Minor, long-term side effects such as pain and fibrosis following RT were common, while 22% developed grade 3–4 toxicity, the majority involving the skin or subcutaneous tissues.4
While this trial confirmed the role of adjuvant RT in reducing the risk of LN field recurrence, the study was conducted prior to the current era where adjuvant immunotherapy and BRAF/MEK inhibiting targeted therapy are available. Unfortunately, the landmark clinical trials of adjuvant systemic therapy do not report factors known to be associated with risk of locoregional recurrence in detail, particularly the presence of ENE, limiting the ability to directly compare the impact of adjuvant RT and systemic therapy on the risk of locoregional relapse.
Furthermore, historically, the standard approach to patients with melanoma with involvement of the sentinel LN (SLN) was to proceed to completion of LN dissection (CLND). However, two trials have since demonstrated no improvement in patient survival in those who proceed to CLND after a positive SLN, and this approach is therefore no longer routinely recommended.5–7
As such, there is currently no data on the role of adjuvant RT in patients with melanoma who have resected locoregional recurrences during or after adjuvant immunotherapy, including those that may or may not have undergone prior CLND of the nodal basin.
Our study aims to answer this question, examining the role of adjuvant RT in patients with melanoma with resected locoregional recurrence during or after adjuvant immunotherapy. Specifically, we sought to determine the rate of locoregional disease control provided by adjuvant RT in this population.
Methods
Patient and study design
The following inclusion criteria were used to retrospectively identify patients for this study: those with resected stage III melanoma (per American Joint Committee on Cancer (AJCC) eighth edition) who received adjuvant anti-programmed cell death protein-1 (PD-1) (±ipilimumab) immunotherapy; locoregional disease recurrence (LN and/or in-transit metastases (ITM)) without distant metastases during or after this immunotherapy; resection of macroscopic disease with or without adjuvant RT at the discretion of the treating multidisciplinary team.
Data on patient, disease and treatment characteristics were collected, including: patient demographics (age, sex, treating center); baseline melanoma features (primary site, thickness, ulceration, mitosis, mutation(s), stage III substage per AJCC eighth edition, performance of CLND, number, site and size of LN(s) involved); adjuvant systemic therapy (drug, duration, reason for cessation) and adjuvant RT (site, dose, toxicity and grade (per National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) V.5.0)); first recurrence details (Eastern Cooperative Oncology Group (ECOG) performance status, lactate dehydrogenase (LDH) concentration, site (LN and/or ITM) and stage of recurrence, method of recurrence detection) as well as treatment (surgery, systemic therapy and/or RT) at first recurrence and response. If adjuvant RT at first recurrence was administered, details on-site, dose, toxicity and grade were collected. Data was captured for up to three recurrences after adjuvant immunotherapy. Patients were followed until death or data censorship date, whichever occurred first. Institutional ethics board approval was obtained for the analysis.
Efficacy assessment
The primary outcome measured was rate of subsequent locoregional recurrence, with secondary outcome measures of locoregional recurrence-free survival (lr-RFS2) and overall RFS (RFS2) to second recurrence. Time to first recurrence was calculated from date of surgery for initial stage III disease to the date of radiological or clinical first recurrence. Lr-RFS2 was calculated from date of resection of first recurrence to date of locoregional radiological or clinical second recurrence. RFS2 was calculated from date of resection of first recurrence to date of any radiological or clinical second recurrence, death or last follow-up. Distant metastasis-free survival (DMFS) was calculated from date of resection of first recurrence to date of distant radiological or clinical second recurrence. OS was calculated from date of resection of first recurrence to date of death from any cause or last follow-up.
Statistical analysis
Categorical variables are summarized as frequencies and percentages, continuous variables are described using median and range. Pearson’s χ2 test and/or Fisher’s exact test was used for comparisons between categorical variables as appropriate. For continuous variables, Wilcoxon rank test was used. Univariable and multivariable logistic regression analyses were conducted to investigate factors associated with rate of locoregional disease recurrence at second recurrence. RFS was analyzed using univariable and multivariable Cox proportional hazard regression. RFS was described using the Kaplan-Meier method and examined by receipt of adjuvant RT at first recurrence. Median survival time and its 95% CI were reported. Differences between survival curves were assessed using a log-rank test. A p value of <0.05 was considered statistically significant. All statistical analyses were conducted using SAS V.9.4 and R V.3.6.0.
Results
Patient, disease and adjuvant treatment characteristics
A total of 71 patients from nine sites in Australia, Europe and the USA were included in the final analysis.
Median age at diagnosis of initial stage III melanoma was 60 years (range, 22–88), 42 (59%) were men, 61 (86%) had cutaneous melanoma, 30 (42%) were BRAF V600E mutant, 34 (48%) were from Australia and 43 (61%) had stage IIIC disease. At initial stage III diagnosis, 47 (66%) had LN only disease and 12 (17%) had both LN and in-transit disease; 38 (74%) had microscopic LN disease identified via SLN biopsy. At resection, CLND was performed in 37 (52%) patients, of which 20 (54%) were for macroscopic disease and 16 (43%) were of the axilla; median number of involved LNs was 1 (0–38), 9 (24%) had ENE and median size of largest LN was 7.6 mm (0–70). Baseline patient and disease characteristics are summarized in table 1.
Patient and disease characteristics at initial diagnosis of resected stage III melanoma
Of the total, 64 (90%) patients received anti-PD-1 monotherapy as adjuvant treatment after resection of initial stage III disease, while 6 (8%) received nivolumab in combination with ipilimumab given at 1 mg/kg every 6 weeks (CheckMate-915 trial8). Median duration of adjuvant therapy was 5 months (1–13). A total of 49 (69%) patients developed recurrent melanoma while still receiving adjuvant immunotherapy. Three patients received adjuvant RT after resection of initial stage III disease, prior to first recurrence. Details of adjuvant therapy are summarized in online supplemental file 1. Median follow-up for all patients was 22 months.
Supplemental material
First recurrence characteristics and adjuvant RT
A total of 49 (69%) patients developed locoregional recurrent melanoma (without distant metastases) while still receiving adjuvant immunotherapy, while 22 (31%) developed recurrence after ceasing adjuvant immunotherapy. For all patients, median time to first recurrence from date of surgery for initial stage III disease was 7 months (1–44). At first recurrence, 58 patients (82%) had an ECOG of 0 and 40 (56%) had a normal LDH concentration. Melanoma stage at first recurrence was stage IIIC in 55 (77%) patients (based on pathological stage at time of recurrence), and recurrence was detected on clinical examination in 39 (55%) patients. Of the 22 (31%) patients that recurred after ceasing adjuvant immunotherapy, 10 (46%) recurred within 3 months of cessation, 6 (27%) recurred within 3–6 months and 6 (27%) recurred after 6 months of cessation.
Most surgical resections at first recurrence resulted in clear margins (n=63, 89%). The majority (n=38, 54%) of locoregional recurrences involved in-transit only disease. Of the 33 patients with locoregional LN involvement at first recurrence, 39% involved axillary LNs, median number of involved LNs was 1 (1–6), 48% had ENE, and median size of largest LN was 15 mm (4–60). Of the three patients that had received adjuvant RT at diagnosis, two patients developed recurrence within the previously irradiated field. Patient, disease and treatment characteristics at first recurrence are summarized in table 2.
Patient, disease and treatment characteristics at first recurrence
After resection of the first recurrence, 24 (34%) patients received adjuvant RT and 47 (66%) did not receive adjuvant RT. Patients who received adjuvant RT had a higher frequency of poor prognostic markers including involved surgical margins (25% vs 4%, p=0.01), ENE (61% vs 33%, p=0.28) and stage IIID disease (17% vs 6%, p=0.38). Furthermore, those who recurred during adjuvant immunotherapy were more likely to receive adjuvant RT (75% vs 66%, p=0.61). Importantly, 75% of patients who received adjuvant RT at first recurrence had nodal (±in-transit) disease, compared with only 32% in the group that did not receive adjuvant RT (p=0.0016). No patient with acral melanoma received adjuvant RT. The majority (54%) of patients who received adjuvant RT at first recurrence did not commence or continue systemic therapy at recurrence, whereas the majority (44%) of patients that did not receive adjuvant RT commenced new systemic therapy or restarted anti-PD-1 therapy. The most common new systemic therapy was adjuvant BRAF/MEK inhibition in both patient groups (table 2). Of the 23 patients who commenced adjuvant BRAF/MEK inhibition 4 (17%) ceased for toxicity (1 of 7 (14%) of those who were also treated with adjuvant RT and 3 of 16 (19%) of those who were not).
Of the 33 patients with locoregional LN involvement at first recurrence, 23 (70%) patients were considered high risk for subsequent locoregional recurrence, as per previously established ANZMTG/TROG criteria where high-risk patients are those with: at least one parotid, two cervical or axillary or three inguinal LNs; maximum size of involved LNs (≥3 cm cervical or ≥4 cm axillary or inguinal LN); or the presence of ENE (table 2).4 Of these 23 patients, 14 received adjuvant RT at first recurrence (p=0.0021).
Of the 24 patients who received adjuvant RT at first recurrence, site of administration of adjuvant RT included: local LN in 18 (75%) patients, primary site in 4 (17%), in-transit disease in 1 (4%) and both primary site and local LN in 1 (4%) patient. Median dose was 48 Gy (range, 36–60) in 20 fractions (range, 6–30). All but one patient was treated with standard RT, with a single patient treated with a hypofractionated schedule (36 Gy in six fractions), no patients were treated with stereotactic dosing. Of the 24 patients who received adjuvant RT, 16 (67%) developed toxicity to RT, of which 15 had dermatitis (±other symptoms) and 1 developed esophagitis. Of the 16 with dermatitis, 10 patients had CTCAE grade 1 and 6 patients had CTCAE grade 2 dermatitis only. The median amount of time required for toxicity to resolve was 34 days (15–288) (online supplemental file 2).
Second recurrence
At a median follow-up from first recurrence of 15 months (95% CI 11 to 20), 33 (46%) developed a second recurrence. Median time from first recurrence to second recurrence was 5 months (1–22). At second recurrence, 23 patients (70%) had an ECOG of 0 and 12 (36%) had a normal LDH concentration. Melanoma stage at second recurrence was IIIC in 18 (55%) patients, and recurrence was detected on imaging in 15 (46%), clinical examination in 13 (39%) and symptoms in 5 (15%) patients. Patient, disease and treatment characteristics at second recurrence are summarized in table 3.
Patient, disease and treatment characteristics at second recurrence
Of the 24 patients that received RT at first recurrence, 7 (29%) developed a second recurrence. Of these 7 patients, 2 (8%) developed locoregional disease while 5 (20%) developed distant disease (table 3). Both patients with locoregional disease had ITM only. Of the 47 patients that did not receive RT at first recurrence, 26 (55%) patients developed a second recurrence, of which 17 (36%) were locoregional and 9 (19%) were distant disease. Of the 17 with locoregional disease, 6 had LN (±in-transit) disease while 11 patients had ITM. Thus, rate of locoregional relapse at second recurrence was 8% (2/24) in those who received adjuvant RT at first recurrence, compared with 36% (17/47) for those who did not receive adjuvant RT at first recurrence, (p=0.01), and there was no difference in rate of distant recurrence (20% vs 19%). Of the 26 patients that did not receive adjuvant RT at first recurrence and developed a second recurrence, 11 (42%) started a new systemic therapy agent. Of the 7 patients that received adjuvant RT at first recurrence and developed a second recurrence, 5 (71%) started a new systemic therapy agent (table 3).
lr-RFS2
Of the 33 patients that developed a second recurrence, 19 patients had locoregional disease. Of these 19 patients, 4 (21%) developed recurrence in the field of prior adjuvant RT (administered either at initial stage III resection or at first recurrence). Median lr-RFS to second recurrence (lr-RFS2) for the total population was not reached (figure 1A). Landmark lr-RFS2 was 85% (95% CI 77% to 93%) at 6 months and 76% (95% CI 67% to 87%) at 12 months. Lr-RFS2 was significantly associated with receipt of adjuvant RT at first recurrence (p=0.014, figure 1B). Landmark lr-RFS at 6, 12 and 18 months was higher for those who received adjuvant RT at first recurrence at 96% (95% CI 88% to 100%), 92% (95% CI 81% to 100%), and 92% (95% CI 81% to 100%), respectively, compared with 79% (95% CI 68% to 91%), 68% (95% CI 56% to 83%), and 63% (95% CI 51% to 79%) for those who did not receive adjuvant RT at first recurrence. On both univariable and multivariable Cox regression analysis, adjuvant RT at first recurrence was associated with longer lr-RFS2 (HR 0.20, p=0.031 and HR 0.16, p=0.015, respectively) (table 4). Commencement of new systemic therapy at first recurrence was also associated with lower likelihood of a locoregional second recurrence on both univariable (HR 0.27, p=0.029) and multivariable (HR 0.21, p=0.012) analysis. Lr-RFS2 was not significantly associated with other factors such as sex, ECOG, LDH and stage at first recurrence (table 4).
Locoregional recurrence-free survival to second recurrence for (A) all patients and (B) by receipt of adjuvant RT at first recurrence (p=0.014). RT, radiotherapy.
Univariable and multivariable Cox regression analysis for locoregional recurrence-free survival at second recurrence
RFS2
For the total population, median RFS after first recurrence (RFS2) was 67 months (95% CI 10 to NR) (figure 2A). Landmark RFS2 was 69% (95% CI 59% to 81%) at 6 months and 57% (95% CI 47% to 70%) at 12 months. While there was a trend towards longer RFS2 for patients who received adjuvant RT at first recurrence, this was not statistically significant (HR 0.54, p=0.072, figure 2B). On both univariable and multivariable Cox regression analysis, commencement of new systemic therapy at first recurrence was associated with a significantly longer RFS2 (HR 0.35, p=0.008, and HR 0.36, p=0.010, respectively). A primary melanoma of non-cutaneous origin was associated with significantly worse RFS2 on both univariable (HR 3.99, p=0.0009) and multivariable (HR 4.64, p=0.004) analysis, though this was likely influenced by small patient numbers (online supplemental file 3).
Recurrence free-survival to second recurrence for (A) all patients and (B) by receipt of adjuvant RT at first recurrence (p=0.083). RT, radiotherapy.
Of the total cohort, six patients died, four due to melanoma progression. Of the remaining two, one patient died from a spinal cord infarction causing diaphragmatic weakness and one patient died from a second primary melanoma that led to malignant ascites; both deaths occurred before index melanoma recurrence.
DMFS
For the total population, the DMFS rate was 83% (95% CI 75% to 92%) at 6 months and 76% (95% CI 68% to 88%) at 12 months. No significant differences were noted in risk of distant relapse between patients who did or did not receive adjuvant RT at first recurrence (online supplemental file 4).
Median OS for all patients was not reached and no significant differences in OS were noted between patients who did or did not receive adjuvant RT at first recurrence.
Propensity score matching
To further explore differences in outcomes between patients who did and did not receive adjuvant RT at first recurrence, we performed a propensity score analysis to mitigate the effect of variables that may have differed between these two groups. A population of 42 patients was identified after matching for site of locoregional disease and disease burden at first recurrence. After matching, the HR for lr-RFS2 for receipt of adjuvant RT at first recurrence compared with no adjuvant RT was 0.18 (p=0.015) (online supplemental files 5 and 6) while the HR for RFS2 was 0.53 (p=0.17) (online supplemental files 5B and 7).
Subgroup analysis
To further determine the efficacy of adjuvant RT at first recurrence, we performed a subgroup analysis where patients who received adjuvant BRAF/MEK inhibitors, ipilimumab monotherapy or anti-PD-1/ipilimumab combination therapy at first recurrence were excluded, as these therapies have a proven RFS benefit. This resulted in a subgroup of 46 patients remaining. Median lr-RFS2 for the subgroup was not reached. The HR for lr-RFS2 for receipt of adjuvant RT at first recurrence compared with no adjuvant RT was 0.17 (p=0.0089) (online supplemental files 8 and 9). Median RFS2 for the subgroup was 12 months (95% CI 6 to NR). The HR for RFS2 for receipt of adjuvant RT at first recurrence compared with no adjuvant RT was 0.32 (p=0.0061) (online supplemental files 10 and 11).
Discussion
To our knowledge, this multicenter, international cohort study is the first study to investigate the role of adjuvant RT in patients with melanoma with locoregional disease recurrence during or after adjuvant anti-PD-1-based immunotherapy. While most multidisciplinary teams would not recommend adjuvant RT in the setting of adjuvant systemic therapy, this is not based on high level evidence, and the role of adjuvant RT when adjuvant systemic therapy fails to control disease is unknown.
We found that adjuvant RT administered after resection of a recurrence that occurred during or after adjuvant immunotherapy significantly reduced the rate of locoregional disease at second recurrence. This was associated with a significant improvement in lr-RFS2 on both univariable (HR 0.20, p=0.031) and multivariable analysis (HR 0.16, p=0.015). This improvement remained significant with propensity score analysis (HR 0.18, p=0.015), conducted to adjust for potential selection bias.
While there was a trend towards improved overall RFS with receipt of adjuvant RT at first recurrence, this was not statistically significant (HR 0.54, p=0.072), though this was likely influenced by small patient numbers. Recognizing the limitations of a retrospective analysis and indirect comparison between retrospective data sets and prospective clinical trials, this result is similar to findings from the seminal randomized ANZMTG/TROG trial, where no significant difference in RFS was noted between those who received adjuvant RT versus observation (HR 0.89, p=0.51).4 As expected, adjuvant RT did not significantly impact the risk of distant relapse. This is similar to the findings from the ANZMTG/TROG trial, where no significant difference was noted in risk of distant relapse between those who received adjuvant RT versus observation (HR 1.07, p=0.73).4
Receipt of a new drug as adjuvant systemic therapy at first recurrence also significantly improved lr-RFS2 on both univariable (HR 0.27, p=0.029) and multivariable analysis (HR 0.21, p=0.012), demonstrating that in some patients, commencing adjuvant BRAF/MEK inhibitors or anti-PD-1/ipilimumab combination therapy represents a suitable option for locoregional disease control. Indeed, other studies have demonstrated the efficacy of these agents as ‘second adjuvant’ therapy, regardless of the receipt of RT.9 Importantly, we undertook a subgroup analysis excluding those patients who received an adjuvant systemic therapy at first recurrence that is known to have an RFS benefit, to account for any confounding effects these active treatments may have on outcomes. This subgroup analysis revealed that receipt of adjuvant RT at first recurrence not only continued to result in a significant improvement in lr-RFS (HR=0.17, p=0.0089), but also led to a significant improvement in RFS (HR=0.32, p=0.0061).
Toxicity from RT was relatively common, with 67% of patients developing some form of reaction to RT. However, the majority (63%) of toxicity was CTCAE grade 1 dermatitis, with only 6 (37%) patients developing grade 2 toxicity and no patients developing grade 3 or higher toxicity. These results compare favorably to the toxicity data from the landmark ANZMTG/TROG trial, where 74% of patients developed grade 2–4 toxicities.4 The reduced rate of toxicity in our study may reflect less accurate recording of toxicity information in a retrospective study and/or an improvement in RT delivery since the randomized trial. Longitudinal data on quality of life and limb volume from the time of first recurrence were not available in this analysis.
Limitations of this study include the retrospective nature of data collection and associated biases. The sample size was relatively small, which may have influenced some results, especially subgroups with small patient numbers (eg, cutaneous vs non-cutaneous primary site). A high rate of CLNDs (52%) was performed at initial diagnosis, and median duration of adjuvant therapy was relatively short (5 months), reflecting a selection bias for patients with high-risk disease in this study. Also, the inclusion criteria for this study required patients to undergo resection of disease at first recurrence, and thus excluded those that received upfront systemic and/or RT at first recurrence. Furthermore, a selection bias for administration of adjuvant RT at first recurrence is evident, favoring those with LN recurrence over those with ITM. A number of patients had data missing on LDH, as can occur in retrospective analyses. Heterogeneity of the study population including differences in characteristics between the two groups (adjuvant RT at first recurrence vs no adjuvant RT at first recurrence) and variations in local practice may have impacted results. Strengths of this study include the inclusion of patients from a variety of countries and each participating site providing data on both patients who did and did not receive adjuvant RT at first recurrence.
Conclusions
This study suggests that adjuvant RT continues to provide control of locoregional disease in today’s modern era, where patients may develop resectable locoregionally recurrent disease during or after adjuvant immunotherapy. Use of adjuvant RT was associated with a lower rate of further locoregional relapse and improved lr-RFS and potentially RFS when accounting for confounding factors such as change of systemic therapy. As seen in previous studies, adjuvant RT does not seem to impact DMFS nor OS. Larger, prospective studies are required to validate these results.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This project has been approved by the Human Research Ethics Committee from MIA & Royal Prince Alfred Hospital (Protocol X15-0454 & HREC/11/RPAH/444). Participating sites gained HREC approval from their respective institutional boards.
Acknowledgments
This study was presented in part as a poster presentation at the virtual American Society of Clinical Oncology (ASCO) annual meeting 2021. PB: NHMRC postgraduate scholarship, AMRF-LEK postgraduate grant. GLon: NHMRC Investigator Grant and the University of Sydney Medical Foundation. AMM: NHMRC Investigator Grant, Nicholas and Helen Moore and Melanoma Institute Australia.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @SerineLo
Contributors Study concept and design: PB, MSC. Data collection: PB, RJ, JM, DBJ, OD, H-LY, GLod, AK, KK. Quality control of data and algorithms: PB. Data analysis and interpretation: PB, AHo, SNL, WW, MSC. Statistical analysis: PB, SNL. Manuscript preparation: PB, MSC. Manuscript editing: PB, MSC. All authors revised the manuscript and approved the submission. All authors are agreeable to be accountable for all aspects of the submitted work. Guarantor: MSC.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests PB: sponsorship: Bristol-Myers Squibb, Merck Sharp & Dohme, Novartis; paid speaker: Novartis, Merck Sharp & Dohme. AH: advisory board member Qbiotics, Oncobeta. JM: consultant advisor: Merck/Pfizer, Merck Sharp & Dohme, Amgen, Novartis, Sanofi, Bristol-Myers Squibb and Pierre Fabre; travel support: Ultrasun, L’oreal, Merck Sharp & Dohme, Bristol-Myers Squibb, Pierre Fabre. DBJ: advisory boards/consultant: BMS, Catalyst Biopharma, Iovance, Janssen, Mallinckrodt, Merck, Mosaic ImmunoEngineering, Novartis, Oncosec, Pfizer, and Targovax. Research funding: BMS and Incyte. AHay: advisory board member for Amgen, Bristol-Myers Squibb, Merck Sharp & Dohme, Novartis, Pierre-Fabre, and Qbiotics. GLod: travel/accommodation/expenses: Sun Pharma. EL: honoraria: Novartis, Medac, Bristol-Myers Squibb, Sanofi, Sun Pharm, Pierre Fabre; consulting/advisory role: Bristol-Myers Squibb, Pierre Fabre and Novartis; travel/accommodations/expenses: Pierre Fabre, Bristol-Myers Squibb, Medac, Sun Pharma. KK: consultant or/and honoraria: Amgen, Roche, Bristol-Myers Squibb, Merck Sharp & Dohme, Pierre Fabre, Novartis; travel support: Amgen, Merck Sharp & Dohme, Bristol-Myers Squibb, Amgen, Pierre Fabre, Medac, Novartis. AHau: consultancy and/or speaker honoraria: Almirall, Amgen, Roche, Bristol-Myers Squibb (BMS), Dermagnostix, Eisai, Immunocore, Merck Sharp & Dohme (MSD), MerckPfizer, Novartis, Philogen, Pierre Fabre, Regeneron, Sanofi, Seagen, SunPharma, Xenthera. GAM: reimbursement of trials costs to the Peter MacCallum Cancer Centre: Array/Pfizer Roche/Genentech; non-reimbursed advisor: Novartis, Bristol-Myers Squibb. AMM: advisory board: Bristol-Myers Squibb, Merck Sharp & Dohme, Novartis, Roche, Pierre Fabre, QBiotics. GLon: advisory board: Agenus, Amgen, Array Biopharma, AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Evaxion, Hexal AG (Sandoz Company), Highlight Therapeutics S.L., Innovent Biologics USA, Merck Sharp & Dohme, Novartis, OncoSec, PHMR, Pierre Fabre, Provectus, Qbiotics, Regeneron. MSC: consultant advisor: Amgen, Bristol-Myers Squibb, Eisai, Ideaya, Merck Sharp & Dohme, Nektar, Novartis, Oncosec, Pierre Fabre, Qbiotics, Regeneron, Roche; honoraria: Bristol-Myers Squibb, Merck Sharp & Dohme, Novartis. All other authors declare no COI.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.