Mismatch repair (MMR)-deficiency and consequently high DNA microsatellite instability (MSI-H) are associated with high tumor mutational burden. A high mutational load increases the potential number of neoantigens that can be presented by the tumor cell and recognized by host lymphocytes. Detection of MMR deficiency and/or MSI-H was proposed as a predictor of an immunogenic tumor and response to anti–programmed cell death protein 1 (PD-1) inhibition, which was confirmed by a landmark study reported by Le et al in 2015.1 Pembrolizumab is currently approved for treatment of MMR-deficient/MSI-H advanced treatment-refractory malignancies regardless of histology.
Michael Cecchini, MD
Mario Sznol, MD
Z1D Subprotocol and KEYNOTE-158
The arm Z1D subprotocol of NCI-MATCH (EAY131) reported by Azad et al and reviewed in this issue of The ASCO Post was a phase II trial of nivolumab in 42 evaluable patients with advanced or metastatic MMR-deficient/MSI-H noncolorectal tumors.2 The results were similar to those obtained in a larger cohort of 233 patients treated with pembrolizumab in KEYNOTE-158.3 Both nivolumab and pembrolizumab block PD-1 and have produced similar results across multiple malignancies. The rationale to duplicate studies with each agent could be questioned in a resource-limited world, but trials such as the one by Azad et al also provide opportunities to conduct correlative studies to develop biomarkers for improved patient selection and to understand the mechanisms of resistance.
Potential Predictive Biomarkers
The Azad et al phase II study and the KEYNOTE-158 trial showed a substantial clinical benefit in a subset of patients treated with anti–PD-1 therapy.2,3 In selecting patients for the trial, the immunohistochemistry assays for MMR protein deficiency and/or polymerase chain reaction–based assays for MSI-H disease likely missed a small number of patients who could have derived treatment benefit.4 The sensitivity of the immunohistochemistry assay was only in the range of 80%, and the results of the immunohistochemistry and polymerase chain reaction assays were not completely concordant.5 For example, Azad et al identified two partial responses in patients who were MMR-deficient but had MSI-low or microsatellite-stable disease by the polymerase chain reaction assay. Data on tumor mutational burden in these patients were not available.
If MMR deficiency or MSI-H are simply indicators of high tumor mutational load, a reasonable argument could be made to assess the tumor more directly for high mutational burden, which may also be present based on dysfunction in other DNA repair pathways. Efforts are ongoing to assess various tumor-sequencing technologies for consistency in determining tumor mutational burden and to identify optimal cut points for offering anti–PD-1 treatment.
“Unfortunately, PD-L1 expression has been, at best, a suboptimal predictive biomarker across tumors, but it remains the only feasible clinical assay available to date.”— Michael Cecchini, MD, and Mario Sznol, MD
Tweet this quote
Responses to anti–PD-1 treatment can also occur even in the absence of high tumor mutational burden. To capture the larger pool of patients potentially responsive to anti–PD-1 therapy, it may be necessary to assess the most downstream determinants of response to anti–PD-1 treatment, such as the presence of tumor-reactive lymphocytes within the tumor microenvironment or draining lymph nodes. Expression of programmed cell death ligand 1 (PD-L1) by tumor or tumor-infiltrating immune cells has been used as a surrogate of an ongoing antitumor immune response that could be activated by PD-1/PD-L1 blockade. Unfortunately, PD-L1 expression has been, at best, a suboptimal predictive biomarker across tumors, but it remains the only feasible clinical assay available to date. At least in some settings, the predictive value of tumor PD-L1 expression (or a tumor T-cell gene activation signature) appears to be independent of tumor mutational burden.
Upfront and Acquired Resistance
Both the KEYNOTE-158 and the arm Z1D subprotocol of NCI-MATCH supported the use of anti–PD-1 as monotherapy only in patients experiencing disease progression after approved standard of care, most often chemotherapy. The durability of response, impressive survival, and low toxicity reported with nivolumab and pembrolizumab in MMR-deficient/MSI-H tumors suggest leveraging these treatments earlier in the treatment plan, either prior to or in combination with the standard of care. Although prospective clinical trials are required, the results of anti–PD-1 trials are sufficiently compelling for clinicians to consider anti–PD-1 treatment prior to the standard of care for an individual patient, depending on the risk/benefit of each treatment.
Despite the impressive activity of anti–PD-1 treatment for MMR-deficient/MSI-H tumors, most patients do not respond, and acquired resistance develops in a subset of responders. Consequently, a better understanding of the mechanisms of upfront and acquired resistance is needed in this patient population. For patients who develop acquired resistance to anti–PD-1 treatment, disruptions in antigen presentation have been described, such as mutations in β2‑microglobulin.4 Furthermore, loss of human leukocyte antigen (HLA) class I is frequently seen in MSI-H colorectal cancer and is likely an important mechanism of upfront resistance to anti–PD-1 treatment.6 Strikingly, early data suggest some malignancies with MMR-deficiency/MSI-H demonstrate much lower rates of response to anti–PD-1—for example, pancreatic cancer and brain tumors.3 The variations in response offer opportunities to define unique mechanisms of resistance, which could then be employed to improve biomarkers for patient selection and develop novel immune-based therapeutic strategies.
“Anti–PD-1 therapy with nivolumab or pembrolizumab should be offered to all patients with treatment-refractory MMR-deficient/MSI-H malignancies….”— Michael Cecchini, MD, and Mario Sznol, MD
Tweet this quote
The combination of dual checkpoint blockade with nivolumab and the anti–cytotoxic T-lymphocyte–associate protein 4 inhibitor ipilimumab produced a 55% objective response rate in MMR-deficient colorectal cancer in CheckMate 142, and subsequently the combination received U.S. Food and Drug Administration approval for MMR-deficient colorectal cancer.7 In the nivolumab monotherapy arm of CheckMate 142, a response rate of 31% was reported in a similar patient population with MMR-deficient colorectal cancer.8 Moreover, responses to ipilimumab/nivolumab were seen regardless of PD-L1 expression, BRAF/KRAS mutations, and Lynch syndrome vs sporadic MMR-deficiency/MSI-H disease.7 In the nivolumab/ipilimumab arm, 71% of patients were progression-free at 12 months.7
The results of CheckMate 142 suggest that the combination of nivolumab and ipilimumab may be able to overcome some of the resistance mechanisms observed with single-agent anti–PD-1 treatment in MMR-deficient/MSI-H malignancies; however, randomized studies against anti–PD-1 alone have not been conducted. Both ipilimumab and ipilimumab/nivolumab demonstrate activity in melanoma progressing after single-agent anti–PD-1 treatment; similar trials could be conducted in patients with MMR-deficient/MSI-H malignancies.
In summary, nivolumab is an effective therapy for treatment-refractory MMR-deficient/MSI-H cancers regardless of tumor type, although activity may vary by tumor type. In arm Z1D subprotocol of NCI-MATCH, the incidence of MMR-deficiency/MSI-H disease varied, but overall was just 2%; although rare, given the profound therapeutic implications, MMR deficiency should be evaluated in all tumors. Anti–PD-1 therapy with either nivolumab or pembrolizumab should be offered to all patients with treatment-refractory MMR-deficient/MSI-H malignancies and should even be considered ahead of standard of care in selected patients. Combinations with dual checkpoint inhibitors such as ipilimumab and nivolumab should be evaluated prospectively in the first-line setting and after disease progression on anti–PD-1 treatment alone.
Furthermore, additional investigation of biomarkers that coexist with MMR deficiency/MSI-H disease is warranted, including mechanisms of acquired resistance such as disruption of antigen presentation. Because defective antigen presentation could be a frequent mechanism of resistance, patients with MMR-deficient/MSI-H malignancies who experience disease progression on anti–PD-1 treatment may be ideal candidates for trials to investigate agents inducing non–T-cell–mediated immunity.
DISCLOSURE: Dr. Cecchini owns stock or other ownership interests in Parthenon Therapeutics; has received honoraria from Agios, AstraZeneca, and Eisai; and has been reimbursed for travel, accommodations, or other expenses by AstraZeneca and Eisai. Dr. Sznol owns stock or other ownership interests in Actym Therapeutics, Adaptive Biotechnologies, Amphivena, Intensity Therapeutics, Nextcure, and Torque; has served as a consultant or advisor to AbbVie, Adaptimmune, Allakos, Almac Diagnostics, Anaeropharma, Array BioPharma, AstraZeneca/MedImmune, Biodesix, Bristol-Myers Squibb, Celldex, Chugai/Roche, Genentech/Roche, Genmab, Genocea Biosciences, GI Innovation, Hinge, Incyte, Immunocore, Innate Pharma, Inovio Pharmaceuticals, Kyowa Hakko Kirin, Lilly, Merck Sharp & Dohme, Modulate, Molecular Partners, Nektar, NewLink Genetics, Gritstone Oncology, Novartis, Omniox, Pieris Pharmaceuticals, Pierre Fabre, Seattle Genetics, Symphogen, Theravance, Torque, and Zelluna; and has held other relationships with AcademicCME, CEC Oncology, Clinical Care Options, Dava Oncology, Haymarket Media, Imedex, Physician Education Resource, Prime Oncology, Research to Practice, TRM Oncology, and Vindico.
1. Le DT, Uram JN, Wang H, et al: PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med 372:2509-2520, 2015.
2. Azad NS, Gray RJ, Overman MJ, et al: Nivolumab is effective in mismatch repair–deficient noncolorectal cancers: Results from arm Z1D—A subprotocol of the NCI-MATCH (EAY131) study. J Clin Oncol 38:214-222, 2020.
3. Marabelle A, Le DT, Ascierto PA, et al: Efficacy of pembrolizumab in patients with noncolorectal high microsatellite instability/mismatch repair–deficient cancer: Results from the phase II KEYNOTE-158 study. J Clin Oncol 38:1-10, 2020.
4. Le DT, Durham JN, Smith KN, et al: Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science 357:409-413, 2017.
5. Bartley AN, Luthra R, Saraiya DS, et al: Identification of cancer patients with Lynch syndrome: Clinically significant discordances and problems in tissue-based mismatch repair testing. Cancer Prev Res 5:320-327, 2012.
6. Kloor M, Becker C, Benner A, et al: Immunoselective pressure and human leukocyte antigen class I antigen machinery defects in microsatellite unstable colorectal cancers. Cancer Res 65:6418-6424, 2005.
7. Overman MJ, Lonardi S, Wong KYM, et al: Durable clinical benefit with nivolumab plus ipilimumab in DNA mismatch repair-deficient/microsatellite instability-high metastatic colorectal cancer. J Clin Oncol 36:773-779, 2018.
8. Overman MJ, McDermott R, Leach JL, et al: Nivolumab in patients with metastatic DNA mismatch repair-deficient or microsatellite instability-high colorectal cancer (CheckMate 142): An open-label, multicentre, phase II study. Lancet Oncol 18:1182-1191, 2017.
In a study (NCI-MATCH trial subprotocol, arm Z1D) reported in the Journal of Clinical Oncology, Nilofer S. Azad, MD, of Johns Hopkins Medicine Sidney Kimmel Comprehensive Cancer, and colleagues found that nivolumab was active in mismatch repair (MMR)-deficient noncolorectal cancers.1