Table 1

Evidence regarding potential biomarkers for tumor-infiltrating lymphocytes (TIL) therapy

BiomarkerStudyTumorFindings
Tumor-related
High TMBLauss et al28; n=27MMSignificant correlation with PFS and OS.
Levi et al29; n=152MMSignificantly higher TMB in ACT responders.
High neoantigen loadLauss et al28; n=27MMSignificant association with PFS and OS.
Creasy et al31; n=64MM
  • Significant association with OS.

  • <200 predicted neoantigens as a potential negative biomarker for response.

Favorable transcriptome and epigenomeLauss et al28; n=25MM
  • Upregulation of immune response-related genes in patients with clinical benefit.

  • High expression of cell-cycle genes in non-responders.

  • CTLA-4, HLA-C, and TAP2 enrichment statistically associated with clinical benefit.

Cascone et al39; n=19MMSignificantly higher expression of glycolytic genes (GPI, PGAM4) in non-responders.
Creasy et al31; n=64MM
  • NGFR, PDE1C, and RTKN2 enrichment associated with improved ORR, PFS, and OS.

  • High ELFN1 expression associated with poor outcomes.

Patient-related
Baseline LDH levelsRosenberg et al7; n=93MMNo significant differences in ORR according to LDH levels.
Besser et al33; n=57MMNegative correlation with response.
Goff et al10; n=101MM
Forget et al32; n=74MMStrong negative effect on OS.
Seitter et al35; n=226MMNegative correlation with response.
Chesney et al34; n=153MM
Rohaan et al13; n=168 (Phase III)MMNo significant differences in PFS (vs ipilimumab) according to LDH levels in a subgroup analysis.
Less prior exposure to systemic therapiesRosenberg et al7; n=93MMNo influence of prior therapies (ipilimumab, IL-2, chemotherapy) on response rates or OS.
Besser et al33; n=57MMPrevious ipilimumab did not correlate with response.
Andersen et al9; n=25MMNo significant difference between responding and not responding patients according to prior therapies (IL-2, ipilimumab).
Goff et al10; n=101MMNo significant difference in response rates according to prior anti-CTLA-4 (n=31); anti-PD-1 (n=3); or both (n=8).
Forget et al32; n=74MMPotential negative impact of previous anti-CTLA-4 on OS was found inconclusive after a multivariable analysis. Non-significant decrease in ORR and DOR in anti-CTLA-4-exposed patients.
Borch et al41; n=55MM
  • Significantly better PFS and DOR in anti-PD-1-naïve patients.

  • Decrease in ORR, CRR, and DOR with increasing number of previous lines.

Seitter et al35; n=226MM
  • Significant decrease in ORR, MSS, and PFS in anti-PD-1 or BRAF/MEKi refractory patients.

  • No impact of prior anti-CTLA-4 therapy on ORR.

Levi et al29; n=181MM
  • Negative impact of previous anti-PD-1 or BRAF/MEKi on ORR.

  • Prior anti-CTLA-4 therapy did not influence outcomes.

Rohaan et al13; n=168 (Phase III)MMNo influence in PFS of prior anti-PD-1 treatment.
Chesney et al34; n=153MMResponses to LN-144 regardless of anti-CTLA-4 exposure (anti-PD-1 refractory population).
Baseline immune-stimulating cytokinesForget et al32; n=74MMHigh baseline IL-9 levels significantly associated with response.
Borch et al41; n=12MMLower mean levels of IL-6, IL-8, IL-10, and TNF in patients with longer PFS.
Treatment and TIL-related
Short ex vivo culture time of TILAebersold et al51; n=67MMSignificantly less days of TIL culture for responders.
Schwartzentruber et al
52; n=41
MMShorter culture duration and TIL doubling time in responders.
Besser et al46; n=20MMLower median TIL age in responders (13.2 days vs 18.6 days).
Itzhaki et al47; n=31MMSignificantly younger TIL in responders (median age 13.2 vs 19.7 days), TIL age at infusion <20 days in all the responders.
Zhou et al54; n=27MMCorrelation between telomere length and clinical responses.
Rosenberg et al7; n=93MM
High total number of TIL infusedBesser et al46; n=20MMHigher number of TIL and higher REP expansion rate in responders.
Ellebaeck et al8; n=6MMMore than 80-fold higher number of CD8+ cells in responders.
Andersen et al9; n=25MMNo significant difference in total number or percentage of CD8+ or CD4+ TIL between responders and non-responders.
Forget et al32; n=74MMHigher number of CD8+ TIL in responders (possibly stronger predictive value in anti-CTLA-4-naïve).
Dafni et al12; n=122 (meta-analysis)MMHigher number of infused cells in responders (11 CR, 36 PR).
Zhou et al21; n=60OsteosarcomaBetter outcomes for cell products with ≥5×109 cells (ORR: 20/35) than those with <5×109 cells (ORR: 2/25).
Itzhaki et al47; n=31MMHigher median number of total TIL and CD8+ TIL in responders.
Radvanyi et al61; n=31MM
Rohaan et al13; n=78 (Phase III)MM
High CD8:CD4 ratioItzhaki et al47; n=31MMHigher percentage of CD8+ T cells in responders (72.1% vs 53%).
Radvanyi et al61; n=31MMHigher percentage of CD4+ TIL in non-responders.
TCR structureAndersen et al9; n=25MMTrend toward a higher number of γδ T cells in non-responders.
Expression of exhaustion markersGros et al79; n=6MMHigher expression of exhaustion markers on CD8+ reactive TIL.
Lauss et al28; n=27MMTrend toward upregulation of exhaustion markers in responders.
Zhou et al21; n=60OsteosarcomaHigher percentage of PD-1+ TIL in non-responders.
Radvanyi et al61; n=31MMHigher percentage of BTLA+CD8+ TIL in responders.
Stoltenborg et al62; n=80MM
Expression of immunosuppressive moleculesRosenberg et al7; n=93MMHigher percentage of CD4+FoxP3+ TIL in non-responders.
Zhou et al21; n=60Osteosarcoma
Yao et al85; n=154MM
Costimulatory signals and phenotypic differentiationHernandez-Chacon et al86; n=NAMMCorrelation between levels of CD137 in CD8+ TIL and cytotoxicity against melanoma cells in vitro.
Rosenberg et al7; n=93MMCorrelation between CD27/CD28 expression and clinical efficacy.
Radvanyi et al61; n=31MMBetter clinical outcomes with CD27− Teff than CD27+ TM cells.
Ye et al87; n=NAOvarian cancerHigher expression of CD137 in TIL than in circulating T cells, and better antitumor effects for CD137+ vs CD137− TIL.
Krishna et al88; n=16MMCorrelation between stem-like CD39−CD69− TIL and responses.
Stoltenborg et al
62; n=80
MMHigher expression of CD27/CD28 and CD39 in TIL from responders.
Broad scope of target antigensCreelan et al15; n=13NSCLCCorrelation between polyclonal T cell reactivity and responses.
Kristensen et al71; n=26MMHigher number of neoepitope-specific infused CD8+ TIL in responders.
Persistence of tumor-reactive cellsRosenberg et al7; n=93MMCorrelation between detection of tumor-reactive CD8+ TIL 1 month after infusion and durable responses.
Forget et al32; n=74MMHigh levels of CD8+ TIL and NK cell-related markers (ULBP-1, NKG2D ligand) 3 months after infusion in responders.
Stevanović et al18; n=29HPV-relatedCorrelation between persistent tumor-reactive TIL and clinical responses.
Creelan et al15; n=13NSCLC
  • ACT, adoptive cell therapy; BRAF/MEKi, BRAF and MEK inhibitors; CRR, complete response rate; CTLA-4, cytotoxic T-lymphocyte-associated protein 4; DOR, duration of response; FoxP3, Forkhead box P3; HPV, human papillomavirus; IL, interleukin; LDH, lactate dehydrogenase; MM, metastatic melanoma; MSS, melanoma-specific survival; NSCLC, non-small cell lung cancer; ORR, objective response rate; OS, overall survival; PD-1, programmed cell death protein-1; PFS, progression-free survival; REP, rapid expansion protocol; Teff, effector T cell; TIL, tumor-infiltrating lymphocyte; TM, memory T cell; TMB, tumor mutation burden; TNF, tumor necrosis factor; ULBP-1, UL16 binding protein 1.