Immune-checkpoint inhibitor therapy, involving anti–CTLA-4 and anti–PD-1/ PD-L1 agents, has transformed the treatment of melanoma. Nonetheless, even with combination anti–PD-1 and anti–CTLA-4 therapy, nearly 50% of patients die of their disease at 5 years.
Immune-checkpoint inhibitor (ICI) therapy, involving anti–CTLA-4 and anti–PD-1/ PD-L1 agents, has transformed the treatment of melanoma. Nonetheless, even with combination anti–PD-1 and anti–CTLA-4 therapy, nearly 50% of patients die of their disease at 5 years.1,2 In patients with advanced, ICI-refractory melanoma (either primary or acquired resistance), the rate of response to retreatment with ICIs or chemotherapy is poor. For example, the overall response rate (ORR) with chemotherapy in this setting is 4% to 10%,1 with a median overall survival (OS) of only 7 to 8 months.3,4
Several abstracts from the recent 2021 American Society of Clinical Oncology Annual Meeting were devoted to the search for better second- and later-line treatment options in patients with advanced melanoma that has progressed on or is refractory to ICIs.
One promising approach in the refractory population is adoptive cell therapy using tumor-infiltrating lymphocytes (TILs) derived from a patient’s own tumor and expanded ex vivo. Lifileucel (Contego), a cryopreserved autologous TIL therapy, has demonstrated efficacy and durable long-term responses in patients with advanced melanoma who progress on or after anti–PD-1 therapy. Results of a global, phase 2, open-label, multicohort, multicenter study (C-14401; NCT02360579) examining the impact of prior anti–PD-1/PD-L1 use on duration of response (DOR) with lifileucel showed that median DOR had not yet been reached,5 as presented by James Larkin, PhD, consultant medical oncologist, The Royal Marsden, NHS Foundation Trust, London, United Kingdom.
Patients in C-144-01 had unresectable or metastatic melanoma treated with at least 1 prior systemic therapy including a PD-1–blocking antibody and, if positive for a BRAF V600 mutation, a BRAF inhibitor with or without a MEK inhibitor.4 Sixty-six patients in the trial were treated with lifileucel, which was infused as a 1-time treatment after nonmyeloablative lymphodepletion. Lifileucel administration was followed by up to 6 doses of IL-2. At enrollment, all 66 patients had received antiPD-1/PD-L1 therapy, 80% had been treated with an anti–CTLA-4 agent, and 52% had received a combination of an anti–PD-1 and anti–CTLA-4 agent.
The ORR was 36.4%; 3 patients had a complete response (CR), and 21 had a partial response (PR). Some 81% of patients had a reduction in tumor burden, and 17.7% had further tumor burden reduction since the April 2020 data cutoff (TABLE 15).
“Responses have continued to deepen over time,” said Larkin during his presentation of the data. “Most of the responses were seen at the time of the first scan.” One PR converted to a CR 24 months after treatment with lifileucel.
The ORR was not predicted by any patient or clinical characteristics assessed, including baseline level of lactate dehydrogenase (LDH), baseline ECOG performance status, and the presence of brain or liver metastases at baseline.
“Early intervention with lifileucel at the time of initial progression on anti–PD-1 agents may maximize benefit,” Larkin said.
Jason J. Luke, MD, director of the Cancer Immunotherapeutics Center, University of Pittsburgh Medical Center, Hillman Cancer Center, Pennsylvania, commented after the presentation that “this is clearly an active regimen in the post–PD-1/CTLA-4 setting. It differentiates mechanistically from other systemic therapies. I would note that the response rate for lifileucel appears to be approximately similar to what we have described previously for pembrolizumab [Keytruda] and low-dose ipilimumab [Yervoy] in the post–PD-1 setting.”
Luke continued, “The duration of response that is reported by these investigators is quite impressive, given that the median has not been reached over a long period of time. Also interesting, it appears that this therapy is predominantly active in those patients who progressed initially on anti–PD-1, and many of those patients had PD-L1–low tumors.”
Percutaneous image-guided cryoablation modulates the immune microenvironment and may also enhance responses in patients with unresectable melanoma who progress on ICIs. As reported in a poster session by Meghan Mooradian, MD, of Massachusetts General Hospital, Boston, and colleagues, a single-center phase 2 study (NCT03290677) examined this approach in 17 patients who had previously received anti–PD-1/ PD-L1 therapy, 12 of whom had primary resistance. Patients underwent cryoablation of an enlarging lesion and continued ICI therapy for at least 2 additional cycles.6
At a median follow-up of 8.5 months, 3 patients had experienced a PR, for an ORR of 18% and a disease control rate (DCR) of 47%. Three patients experienced periprocedural events: 1 grade 2 pneumothorax, 1 grade 2 diaphragm puncture, and 1 grade 3 infection.
“Efficacy of this potentially synergistic approach in patients with PD-1–refractory metastatic melanoma is encouraging,” the investigators concluded. Correlative analyses are in process to identify potential biomarkers of response.
A less successful strategy for recurrence appears to be ICI rechallenge. Alexander M. Eggermont, MD, PhD, chief scientific officer at Princess Máxima Center, Utrecht, the Netherlands, presented data from the EORTC 1325-MG/KEYNOTE-054 study of pembrolizumab in patients with completely resected high-risk stage III melanoma (NCT02362594). Among 20 patients who experienced disease recurrence and were rechallenged with pembrolizumab, the median progression-free survival (PFS) was 4.1 months.7 In a more select group of 9 patients who had stage IV disease and had recurrence more than 6 months after the completion of 1 year of pembrolizumab, the activity of pembrolizumab was low, with only 1 CR (11%).
Several combinations involving immunotherapy are under investigation in the second- line setting.
Domatinostat/pembrolizumab. The combination of domatinostat and pembrolizumab induced an ORR of 7.5% and a DCR of 30% in the SENSITIZE study (NCT03278665) of patients with advanced (unresectable or metastatic) melanoma refractory or unresponsive to prior ICI therapy,8 according to preliminary phase 1b data presented by Jessica C. Hassel, MD, of Heidelberg University Hospital, Germany, and colleagues.
Domatinostat is a class I–selective histone deacetylase inhibitor that influences the tumor microenvironment by increasing the expression, processing, and presentation of tumor-associated antigens, as well as promoting infiltration of cytotoxic T cells into tumors. Patients were enrolled in 5 different domatinostat dose cohorts, ranging from 100 mg/day to 200 mg twice daily, combined with pembrolizumab at 2 mg/kg every 3 weeks.
All patients had stage III or IV melanoma, and approximately one-third had elevated LDH levels.
At baseline, most patients had had an inflammatory tumor phenotype and T-cell exhaustion. Increases in inflammation scores, major histocompatibility complex molecules, and antigen-processing machinery components were observed. “The translational research results support the immune-modulating mode of action to synergize with checkpoint inhibition,” said Hassel.
Ceralasertib/durvalumab. In a phase 2 study (NCT03780608), the investigational ATR kinase inhibitor ceralasertib exhibited activity when combined with durvalumab (Imfinzi) in patients with metastatic melanoma who had confirmed progression on anti–PD-1 therapy,9 reported Minsuk Kwon, MD, PhD, of Samsung Medical Center, Sungkyunkwan University, Seoul, South Korea.
“ATR kinase is one of the central regulators in the DNA damage response [DDR] pathway. Alterations in DDR are associated with increased somatic tumor mutational burden,” said Kwon.
The melanoma study enrolled 30 patients with confirmed progression on an anti–PD-1 agent. Patients with active brain metastases were excluded. Ceralasertib 240 mg was administered twice daily on days 15 to 28, and durvalumab 1500 mg was given on day 1 of a 28-day cycle. Treatment continued until disease progression.
Median PFS was 7.1 months, and median OS was 14.2 months. Best response to the ceralasertib/durvalumab combination was PR in 9 patients and stable disease in 10, for an ORR of 30% and a DCR of 63.3%.
Siwen Hu-Lieskovan, MD, PhD, a discussant during the presentation and director of solid tumor immunotherapy at Huntsman Cancer Institute, University of Utah, Salt Lake City, was impressed with this combination.
Interestingly, a phase 1 trial (NCT02630199) by the same group of investigators showed a similar response rate to ceralasertib combined with chemotherapy,9 she noted. The ORR in the trial was 33%, calling into question the benefit of adding durvalumab, which gave an ORR of 30% (TABLE 210). However, the median PFS and OS were almost doubled with the combination of ceralasertib with durvalumab compared with ceralasertib and chemotherapy, “which seems to suggest that adding durvalumab might increase the survival benefit, even though this is an antiPD-1–resistant population, [perhaps due to] the potential synergistic effect of ceralasertib with durvalumab,” Hu-Lieskovan said during presentation of the data.
Ceralasertib, which is considered a targeted therapy, is usually highly effective at suppressing tumor growth but is limited by a high rate of relapsed disease over time. ICIs such as durvalumab are associated with a lower response rate but prolonged patient survival.
“When we combine a targeted therapy with immunotherapy, the goal is not an additive effect but synergy to achieve a high response rate with prolonged survival in the majority of patients,” said Hu-Lieskovan. The longer median PFS and OS suggest that ceralasertib and durvalumab are synergistic, she said, given that these patients had already progressed on anti–PD-1 agents in the past.
The potential synergy could be due to increased tumor mutational burden from the DNA damage repair defect, an influence on the tumor immune microenvironment, or a direct effect on immune cells, she said.
Nivolumab/dabrafenib/trametinib. The triplet combination of nivolumab (Opdivo) with the mechanisms of the BRAF inhibitor dabrafenib (Tafinlar) and the MEK inhibitor trametinib (Mekinist) induced an impressive 88% ORR in the anti–PD-1–refractory setting in patients with BRAF-mutated metastatic melanoma in the phase 2 TRIDeNT study (NCT02910700),11 said Elizabeth M. Burton, MBA, of The University of Texas MD Anderson Cancer Center in Houston.
“Clinical trials evaluating BRAF-MEK inhibitors in combination with PD-1 [inhibitors] have reported mixed results. Notably, all trials excluded patients previously treated with PD-1 [inhibitors] and also patients with active brain metastases,” said Burton during the presentation. “Additionally, retrospective studies have reported short median PFS of around 5 months for patients treated with targeted therapy after PD-1 [inhibitor] failure,” pointing to the unmet need to identify effective therapies for patients who have active brain metastases and have progressed on immunotherapy.
The single-center TRIDeNT study enrolled 27 patients with PD-1–naive or –refractory BRAF-mutated metastatic melanoma, with or without active brain metastases.
At a median follow-up of 18.4 months, the median PFS was 8.2 months in the 16 evaluable patients who were refractory to anti–PD-1 therapy. The ORR was 92% in the overall study population: 100% in the anti–PD-1 therapy–naive patients (n = 10), with 1 CR, and 88% in the anti–PD-1– refractory cohort, with 2 CRs. Median PFS was not different between those with and without brain metastases. “What is striking with our cohort, albeit with small numbers, is the median PFS of 8 months [in patients with brain metastases],” Burton said.
Lenvatinib/pembrolizumab. The LEAP-004 study (NCT03776136) assessed the combination of lenvatinib (Lenvima) and pembrolizumab in patients with confirmed progression on an ICI. Updated findings after 3 months of additional follow-up (median follow-up, 15.3 months) revealed 1 additional CR, a disease control rate of 66.0%, and an improvement in the median DOR from 6.3 to 8.2 months,12 reported Ana Maria Arance Fernandez, MD, PhD, of Hospital Clinic Barcelona in Spain.
LEAP-004 enrolled 103 patients with progressive disease confirmed per iRECIST within 12 weeks of their last dose of a PD-1/ PD-L1 inhibitor given alone or with antiCTLA-4 or other therapies. Patients received lenvatinib 20 mg once daily plus pembrolizumab 200 mg every 3 weeks until disease progression or unacceptable toxicity.
The ORR was 33.3% in the 30 patients with progression on prior anti–PD-1 plus antiCTLA-4 therapy, 18.2% in the 11 patients who received prior anti–PD-1/PD-L1 treatment in the adjuvant setting only, 22.6% in the 62 patients with primary resistance to prior anti–PD-1/PD-L1 therapy in the advanced setting, and 22.7% in the 22 patients with secondary resistance in the advanced setting.
Lenvatinib plus pembrolizumab “is an active regimen in the [post–]PD-1/CTLA-4 setting,” commented Luke. “It’s unclear to me whether or not this efficacy really differentiates from what we’ve seen from previous combinations.”
The ORR for the entire patient cohort remained 21.4%, which “appears on first pass to be useful in a PD-1–refractory setting, but I’ll note that the duration of response is less than one might expect for checkpoint blockade,” he added. “[The DOR] has improved from 6 months at their initial report to 8.3 months now, but I think, again, this calls into question the mechanism whereby we’re pursuing this anticancer effect.”
Alrizomadlin/pembrolizumab. Another strategy that combined an MDM2 inhibitor with an ICI, was studied in a phase 2 trial (NCT03611868).13 The investigators assessed the combination of pembrolizumab and alrizomadlin, an orally bioavailable MDM2 antagonist.
“MDM2 blocks normal p53 function by binding to the p53 protein, leading to ubiquitination and protein degradation,” said Anthony W. Tolcher, MD, FRCPC, of Texas Oncology-San Antonio Babcock and lead investigator of the study. “Retrospective clinical data indicate that MDM2 amplification in tumors is associated with tumor hyperprogression to checkpoint inhibitors.”
Alrizomadlin (APG-115) restores p53 function, activating p53-mediated apoptosis in tumor cells with wild-type TP53 (gene encoding p53) and/or MDM2 amplification. Preclinical studies demonstrate synergy between alrizomadlin and PD-1 blockade in both wild-type and TP53-mutant tumors in syngeneic murine tumor models.
The phase 2 study included a cohort of 26 patients with melanoma that was resistant to immuno-oncology agents. Other cohorts enrolled patients with resistant non–small cell lung cancer and solid tumors with ATM mutations.
The ORR in the melanoma cohort, of which 29 patients were evaluable for efficacy, was 24.1% and the DCR was 55.2%. Among the responders, there was 1 CR and 6 PRs, with 2 of the PRs being unconfirmed. Results remain preliminary in the other expansion cohorts. In the immuno-oncology–resistant melanoma cohort, the combination appeared to perform best in patients with cutaneous melanoma, with an ORR of 26.7% and a DCR of 46.7%.
What Might the Future Look Like?
An optimal approach might be frontline treatment with an anti–PD-1 therapy, perhaps combined with an anti-LAG3 antibody, said Luke. “[Patients could] then could be treated with PD-1 and low-dose CTLA-4, and then have those TILs available so that they could be reinfused should they progress on that regimen,” he said.
Mario Sznol, MD, of Yale University School of Medicine, New Haven, Connecticut, agrees that that the choice of second-line therapy would depend on the frontline treatment. “It becomes a little complex,” he said in an interview with Targeted Therapies in Oncology. If a LAG3 antibody with an anti–PD-1 agent is used up front, the choice upon progression would depend on whether an anti–CTLA-4 therapy would have activity in a patient already treated with a LAG3 antibody.
Interpreting activity in the resistant/ refractory setting may be difficult because of the potential for late responses to anti–PD-1 therapy and/or anti–CTLA-4 agents, said Sznol. TILs have shown activity in patients with progression on combined anti–PD-1 and anti–CTLA-4 treatment. “If and when TIL [therapy is] approved, and many of us think it will be approved, it would be a reasonable second-line option for patients with melanoma,” he said. “The durability of some of those responses is very encouraging.”
A number of experimental agents being investigated in the second-line setting (some of which are summarized above) appear promising, he said, but “the problem is we don’t have biomarkers, and it’s possible that all those agents would address a small subset of patients. For example, anti-CD40 is one. Anti–IL-8 is another. But it’s possible that when the patient has already had anti–PD-1 and a LAG3 antibody, those other agents may not have a lot of activity. TILs still seem to be active in patients who have had the combination of ipilimumab and nivolumab. It could be that TILs won’t be active in patients who receive an anti–PD-1 therapy and nivolumab and then a LAG3 antibody and nivolumab.”
The use of newer agents empirically in the clinic, perhaps sequentially, could improve 5-year OS rates, said Sznol. “The exciting part is that these are not overlapping subsets of patients, and some way or another, we’re going to figure out who to use them in,” he said. “And at the end of the day, instead of 50% 5-year survival, it will be at 65% or 70%.”