There have been major immunotherapy advancements in the treatment of solid tumors over the last few years—like the two monoclonal antibodies blocking the programmed death-1 (PD-1) receptor, nivolumab (Opdivo) and pembrolizumab (Keytruda), which are now approved by the U.S. Food and Drug Administration for the treatment of renal cell carcinoma, melanoma, and non-small cell lung cancer.
The PD-1 pathway negatively regulates the innate immune TH1 cytotoxic response to cancer cells. Many tumors have upregulation of PD-1 and programmed-death ligand-1 (PD-L1) that enables cancer cells to evade the host immune response. Association of PD-1 and PD-L1 inhibits immune function. Blocking PD-1 was theorized to be a therapeutic strategy to restore anti-tumor immune activity.
Accumulation of Errors
Patients with metastatic colorectal cancers refractory to initial treatments that manifest microsatellite instability (MSI) were recently reported to have durable responses with pembrolizumab. Microsatellites are short repetitive sequences of DNA that occur throughout the genome. MSI is due to DNA mismatch repair deficiency, a condition that leads to an accumulation of errors within and truncation of microsatellites.
Mismatch repair proteins (MLH1, MSH2, MSH6, and PMS2) are responsible for the surveillance and correction of mistakes made by DNA polymerase; thus, when mismatch repair genes are mutated or silenced, mismatch repair deficiency occurs. Analogous to a defective spellcheck in a word processing program, this leads to thousands of mutations in tumor cells that exhibit MSI.
MSI may be due to sporadic mutations leading to hypermethylation of MLH1 promoter or inherited germline mutations in the mismatch repair genes (OT 3/10/16 issue). MSI status is determined by two methods, immunohistochemistry (IHC) and polymerase chain reaction (PCR). Tumors are determined to manifest MSI (also known as MSI-high) if IHC staining shows that one or more of the mismatch repair proteins (MLH1, MSH2, MSH6, and PMS2) are missing or if PCR testing shows greater than 30 percent of the microsatellites studied show instability when compared to normal tissue.
Colorectal cancers that have MSI also have increased tumor infiltrating lymphocytes (TIL) when viewed under the microscope. In addition, they exhibit higher expression of checkpoint ligands such as cytotoxic T lymphocyte-associated-4 (CTLA-4), PD-L1, and PD-1 in TIL, stroma, and invasive front (Cancer Discov2015;5:43-51).
Phase II Study
The fact that pembrolizumab leads to frequent responses and long-lasting extension of progression free survival is distinctive; regorafenib (Stivarga) and TAS-102 (Lonsurf), therapies recently approved for refractory metastatic colorectal cancer, prolong overall survival but are not associated with improvements in objective response rates. Those agents both prolong progression-free survival compared to supportive care alone, but the median extension was less than two months with both compounds.
Dung et al conducted a Phase II study of pembrolizumab in three treatment refractory patient populations: MSI-high colorectal cancer patients, microsatellite stable (MSS) colorectal cancer patients, and non-colorectal MSI-high cancer patients (N Engl J Med2015;372:2509-20). Primary end points were immune-related objective response rate and 20-week immune-related PFS rate.
For the MSI colorectal cancer patients, immune-related overall response rate was 40 percent (4/10 patients; 95% CI 12 to 74) and immune-related progression-free survival rate at 20 weeks was 78 percent (7/9 patients; 95% CI 40 to 97). Patients with MSS colorectal cancer had immune-related objective response rate of 0 percent (95% CI 0 to 20) and the immune-related progression-free survival rate at 20 weeks was 11 percent (2/18 patients; 95% CI 1 to 35).
Patients with MSI-high tumors that did not originate in the colon or rectum had the same magnitude of benefit as did the MSI-high colorectal cancer patients. Treatment was well-tolerated and the most common side effects were rash or pruritus (24%), thyroiditis, hypothyroidism, or hypophysitis (10%), and asymptomatic pancreatitis (10%). CEA response were seen in 7/10 patients with MSI-high colorectal cancer, and occurred prior to radiographic response.
Correlative studies were performed, and an increased number of CD8+ lymphoid cells at the invasive fronts of MSI-high tumors were noted. Membranous PD-L1 expression was only seen in MSI-high tumors, specifically in TILs and tumor-associated macrophages at the invasive fronts. However, CD8 and PD-L1 expression did not correlate with progression-free survival or overall survival, and thus further investigation is needed to see if PD-L1 expression will be a predictive biomarker of response to PD-1 inhibitors.
The finding of a high number of somatic mutations in melanomas has correlated with improved likelihood of response to anti-CTLA-4 therapy, and this may be true in MSI-high colorectal cancer with anti-PD1 therapy. MSI-high tumors have 10- to 100-fold more somatic mutations than MSS tumors, and the mutations may form tumor-specific neoantigens that lead to immune infiltrate at the tumor invasive fronts. Le et al report that there was a mean number of 1782 somatic mutations per tumor in MSI-high tumors and 73 mutations per tumor in patients with MSS tumors (ASCO 2015 Annual MeetingAbstract LBA100). Longer progression-free survival correlated with high numbers of somatic mutations and possible mutation-associated neoantigens, although the number of samples analyzed were small (n=15).
The MSI phenotype in colorectal cancer has therapeutic implications, in addition to identifying patients with Lynch syndrome and affecting surveillance practices. For early stage colon cancers, MSI is indicative of improved patient outcome and lack of benefit of 5FU for Stage II colon cancers. For patients with metastatic colorectal cancer, MSI is likely a predictive biomarker for anti-PD-1 treatment efficacy based on ongoing clinical studies.
Pembrolizumab was recently granted breakthrough status for this patient population by the FDA, and a Phase II study (KEYNOTE-164) is ongoing. There is also a first-line Phase III study of pembrolizumab versus investigator-choice chemotherapy for MSI-high colorectal cancers. However, only 5 percent of metastatic colorectal cancers exhibit MSI; the challenge will be to understand the mechanism by which MSI sensitizes the tumors to PD-1 inhibition and identify ways to stimulate the immune system in MSS colorectal cancers. Clinical trials in progress or on the drawing board will be combining PD-1 inhibitors with chemotherapy, vaccines, and radiation to enhance the immune system.