Tumor mutation burden and circulating tumor DNA in combined CTLA-4 and PD-1 antibody therapy in metastatic melanoma - results of a prospective biomarker study

Abstract

Background: Metastasized or unresectable melanoma has been the first malignant tumor to be successfully treated with checkpoint inhibitors. Nevertheless, about 40-50% of the patients do not respond to these treatments and severe side effects are observed in up to 60%. Therefore, there is a high need to identify reliable biomarkers predicting response. Tumor Mutation Burden (TMB) is a debated predictor for response to checkpoint inhibitors and early measurement of ctDNA can help to detect treatment failure to immunotherapy in selected melanoma patients. However, it has not yet been clarified how TMB and ctDNA can be used to estimate response to combined CTLA-4 and PD-1 antibody therapy in metastatic melanoma.

Patients and methods: In this prospective biomarker study, we included 35 melanoma patients with ipilimumab (anti-CTLA-4) and nivolumab (anti-PD-1) therapy. In all patients, a tumor panel of 710 tumor-associated genes was applied (tumor vs. reference tissue comparison), followed by repetitive liquid biopsies. Cell-free DNA was extracted and at least one driver mutation was monitored. Treatment response was evaluated after about three months of therapy.

Results: TMB was significantly higher in responders than in nonresponders and TMB > 23.1 Mut/Mb (TMB-high) was associated with a survival benefit compared to TMB ≤ 23.1 Mut/Mb (TMB-low or TMB-intermediate). Furthermore, a > 50% decrease of cell-free DNA concentration or undetectable circulating tumor DNA (ctDNA), measured by tumor-specific variant copies/ml of plasma at first follow-up three weeks after treatment initiation were significantly associated with response to combined immunotherapy and improved overall survival, respectively. It is noticeable that no patient with TMB ≤ 23.1 Mut/Mb and detectable or increasing ctDNA at first follow-up responded to immunotherapy.

Conclusion: High TMB, > 50% decrease of cell-free DNA concentration, and undetectable ctDNA at first follow-up seem to be associated with response and overall survival under combined immunotherapy. The evaluation of ctDNA and cell-free DNA three weeks after treatment initiation may be suitable for early assessment of efficacy of immunotherapy.

Conflict of interest statement

AF served as consultant to Roche, Novartis, MSD, Pierre-Fabre; received travel support from Roche, Novartis, BMS, Pierre-Fabre, received speaker fees from Cegat, Roche, Novartis, BMS, MSD. TA received travel support from Novartis. TE served as consultant to Roche, Novartis, MSD, BMS, Pierre-Fabre; received speaker fees from Roche, Novartis, BMS, MSD. CG reports grants and personal fees from Novartis, BMS, Roche, personal fees from MSD. Personal fees from Amgen, Philogen, LEO, Incyte, outside the submitted work. DD received travel support from BMS. No competing interests were declared by the other authors.

Figures

Fig. 1

a Comparison of tumor mutation burden (TMB) in responders and non-responders to combined immunotherapy. b Comparison of tumor mutation burden (TMB) in complete responders, partial responders, and non-responders to combined immunotherapy

Fig. 2

a Cell-free DNA concentrations at start of combined immunotherapy (x axis) and at first follow-up (3–4 weeks later, y axis). Patients were classified into three groups, depending on the change in their cell-free DNA concentration as increasing (increase of more than 50%), decreasing (decrease of more than 50%), or stable. The respective thresholds are marked by broken lines. Increase of cell-free DNA is observed more frequently in non-responders. The four highest values can be seen in the inserted image in the upper right corner. b ctDNA, measured by tumor-specific variant copies/ml of plasma at start of combined immunotherapy (x axis), and at first follow-up (3–4 weeks later, y axis). Increase of ctDNA is almost only observed in progressive patients. Please note that multiple patients had undetectable ctDNA at both time points and are not visible in the plot as separate points due to overplotting (2 for complete response, 4 for partial response, 3 for progress)

Fig. 3

Impact of baseline patients’ and disease characteristics on overall survival since the beginning of combined immunotherapy. 1Log rank test / 2Log rank test for Trend. *significant. a Tumor mutation burden (TMB) > 23.1 Mut/Mb vs. TMB ≤ 23.1 Mut/Mb, p = 0.061. b ctDNA measured by tumor-specific variant copies/ml of plasma detectable vs. undetectable at first follow-up, p = 0.006*1. c ctDNA measured by tumor-specific variant copies/ml of plasma increasing vs. not increasing at first follow-up, p = 0.03*1. d Cell-free DNA decrease > 50% vs. stable vs. increase > 50%, p = 0.005*2. e Targeted treatment (TT) before start of combined immunotherapy vs. no TT before, p = 0.001*1. f Men vs. women, p = 0.005*1. g Liver metastasis baseline vs. no liver metastasis baseline, p=0.013*1. h LDH baseline normal vs. elevated, p = 0.001*1

Fig. 4

Impact of combined variables of TMB on overall survival since the beginning of combined immunotherapy. TMB > 23.1 Mut/Mb (TMB-H) TMB ≤ 23.1 Mut/Mb (TMB-L). Log rank test for Trend. *significant. a Tumor mutation burden (TMB) and ctDNA measured by tumor-specific variant copies/ml of plasma detectable vs. undetectable at first follow-up p = 0.005*. b Tumor mutation burden (TMB) and ctDNA measured by tumor-specific variant copies/ml of plasma increasing vs. not increasing, p = 0.032*. c Tumor mutation burden (TMB) and cell-free DNA decrease > 50% vs. stable vs. increase > 50% at first follow-up, p = 0.016*. d Tumor mutation burden (TMB) and presence of liver metastases, p = 0.018*. e Tumor mutation burden (TMB) and sex, p = 0.010*