Alexander Noor Shoushtari, MD, Memorial Sloan Kettering Cancer Center, New York.
according toTebentafusp is the first therapy to demonstrate an overall survival (OS) benefit in uveal melanoma, Dr. Shoushtari noted in a 2021 European Society of Medical Oncology Congress virtual oral presentation Sept. 17 (abstract 17570). He noted further that, in prior research, OS was improved regardless of RECISTv1.1 best response, suggesting that better surrogate efficacy endpoints are needed.
Uveal melanoma is a rare melanoma type with low mutational burden, but frequent liver metastases. Benefit from immune checkpoint inhibitors is poor, and there is no established standard of care once the disease becomes metastatic. “Immune checkpoint inhibitors are not as good for treating this type of melanoma as they are for treating cutaneous disease, and traditionally preferred treatment is within clinical trials,” Dr. Shoushtari said. In frontline trials, 1-year survival has been in the 50% range. Tebentafusp is an investigational, first-in-class bispecific soluble T-cell receptor (TCR) therapeutic. It is designed to target gp100 (a melanoma-associated antigen) through a high-affinity TCR-binding domain and an anti-CD3 T-cell–engaging domain, which redirects T cells to kill gp100 positive melanocytic expressing tumor cells.
Prior research has demonstrated a disconnect between RECIST response classification and tebentafusp OS benefit. In the IMCgp100-202 study among patients treated first-line for metastatic uveal melanoma with tebentafusp or investigator choice, intent-to-treat analysis showed a survival probability benefit for tebentafusp (hazard ratio, 0.51; 95% CI, 0.37-0.71), with a best response of progressive disease population HR of 0.43 (95% CI, 0.27-0.68). While the RECIST response rate was only 9.1%, the HR for progression-free survival was 0.73 (95% CI, 0.58-0.94). “That suggests that RECIST is not a fantastic way to predict who will benefit from this drug,” Dr. Shoushtari stated.
Similarly in the IMCgp100-102 study of tebentafusp monotherapy in second-line metastatic uveal melanoma (n = 127), the RECIST response rate was 5%. Duration of response was 8.7 months and median OS was 16.8 months. Historical second-line OS has been reported at 7.8 months. The 1- and 2-year survival (62%/37%) compared favorably with historical rates (37%/15%), as well. Dr. Shoushtari noted that 92% of patients had detectable ctDNA with mutations in known uveal melanoma oncogenes. He pointed out that baseline ctDNA levels significantly correlated with tumor burden. Also, 70% of evaluable patients had any ctDNA reduction, with 0.5-3.2 log reduction in 99.9%, a 0.5 log reduction in 68% and 3 log reduction (cleared) in 14% of patients. ctDNA reduction, Dr. Shoushtari said, was associated with greater mean tumor shrinkage and with less tumor growth. Importantly, there was a linear correlation between ctDNA reduction and better OS (R2, 0.88, P < .0001).
Among all evaluable patients, comparing those with less than 0.5 log ctDNA reduction with those with at least a 0.5 log reduction showed a hazard ratio of 0.56 (95% CI, 0.32-0.95; P = .03). Among those whose best response was progressive disease, 35% had at least a 0.5 log reduction in ctDNA with an OS hazard ratio of 0.44 (95% CI, 0.2-0.94; P = .027), compared with less than a 0.5 log reduction. Among those whose best response was stable disease, 28% had at least 1 log reduction with a hazard ratio of 0.48 (95% CI 0.16-1.43; P = .16) for OS, compared with those with less than 1 log reduction. Dr. Shoushtari pointed out that “14% of patients cleared ctDNA, including some (n = 12) with best RECIST responses of stable or progressive disease. All patients with ctDNA clearance were alive beyond 1 year; with a hazard ratio, compared to those who had not cleared ctDNA, of 0.14 (95% CI, 0.03-0.57).
Summing up, Dr. Shoushtari said that ctDNA was detectable in more than 90% of second-line tebentafusp-treated patients with metastatic uveal melanoma and correlated with tumor burden. About 70% had ctDNA reduction versus 39% with tumor shrinkage and 5% RECIST response. The linear correlation between the magnitude of ctDNA reduction and improved OS on tebentafusp, was uncoupled from best RECIST response. “For tebentafusp, ctDNA reduction may be a better surrogate of overall survival than RECIST response.”
The ESMO-appointed discussant for the study, Christian Rolfo, MD, PhD, MBA, Icahan School of Medicine at Mount Sinai, New York, examined the tebentafusp study author’s claim that the radiographic assessment of tumors may underestimate the effect of tebentafusp, compared with ctDNA. The strengths of the study include, he said, that it is a drug- and tumor-specific evaluation of the role of ctDNA as a surrogate of response. “Its strength is that it shows an important correlation between ctDNA levels and overall survival, and that response rate is evaluated better with ctDNA.” A question that remains open, Dr. Rolfo added, is whether RECIST criteria are still a good comparator for biologic response.
The study was funded by Immunocore Dr. Shoushtari disclosed numerous pharmaceutical-related financial interests.
This article was updated Sept. 24, 2021.