What adjuvant chemotherapy would you recommend in an elderly patient with resected stage III colon cancer with MMR deficiency with MLH1 promoter hypermethylation?

This is a great question and actually a quite common situation we see in the clinic. Majority of the dMMR colorectal cancer is actually due to MLH1 promotor hypermethylation (about 75% of all dMMR colorectal cancer) while a small portion of dMMR colorectal cancer is due to Lynch syndrome. These patients are usually elderly female with right sided primary tumor that is poorly differentiated. Braf V600E mutation is a good surrogate biomarker for these sporadic dMMR tumors. We can also check for MLH1 promotor hypermethylation directly to confirm the sporadic nature of the disease without a germline test. 

Since this is stage III dMMR colon cancer, and we have strong evidence that fluoropyrimidine single agent chemotherapy is not effective/beneficial as adjuvant treatment (Jin and Sinicrope, PMID 33467526), there is evidence that oxaliplatin-based adjuvant chemotherapy is effective (actually the benefits seemed to be greater in dMMR stage III colon cancer based on the most recent meta-analysis data from ACCENT database, Cohen et al., PMID 33356421). So we should offer oxaliplatin-based adjuvant chemotherapy for these patients. The IDEA study did not have details regarding the MSI/MMR status so we do not have the data (but this will also depend on the high risk, low risks). The current guidelines do not put dMMR stage III colon cancer as a separate group (not like in the stage II setting specifically emphasizes that adjuvant treatment is not recommended). I would think 3 months of CAPOX will be a very reasonable approach for these patients if they are candidates for adjuvant oxaliplatin-based chemotherapy. 

However, we know that immunotherapy is very effective in metastatic dMMR colorectal cancer and there are also some evidence that immunotherapy could be very effective in early stage colorectal cancer (NICHE study, Chalabi et al., PMID 32251400) even in pMMR colon cancer. So currently, there are two phase III randomized studies evaluating immunotherapy in dMMR stage III colon cancer patients. One is the ATOMIC study (patients will have FOLFOX for 12 cycles with/without atezolizumab, and the experimental arm will have atezolizumab continued as monotherapy for an additional 6 months). The other trial is POLEM which evaluates fluoropyrimidine-based chemotherapy alone or combined with Avelumab. These studies are ongoing, and we have the ATOMIC study available in our clinic, so I have been putting patients on the study. However, this study's chemo regimen is 6 months FOLFOX with 5Fu bolus, so it is quite tough for some elderly patients although the study allows dose adjustment. The studies are well designed but will not be able to answer the question if single agent anti-PD-1 or anti-PD-L1 will be as good as chemoimmunotherapy in the adjuvant setting. We also do not know if the studies have enough statistical power to draw conclusions given the low event rates we see in the real world nowadays. It will be a while before we will have results from these two studies to guide our adjuvant treatment for these patients. 

Meanwhile, for some elderly patients who are not robust or with significant comorbidities, I usually will explain the pros and cons of adjuvant treatment to them. I will also get a ctDNA test for them, if they (or we) are on the fence for adjuvant treatment, then the patients/we are more comfortable with a negative ctDNA result. If the ctDNA result is positive, we may still hold chemo (if we are concerned for side effects/tolerance from the beginning) but I will get more image studies (such as PET or liver MRI) and closer follow-up image studies/blood tests knowing there is no phase III study supporting modifying treatment based on ctDNA result. I personally believe circulating tumor DNA will help individualize our treatment in the near future.

One could also argue, that in an elderly frail patient with a low absolute benefit of adjuvant chemotherapy and its attendant side effects consider surveillance and if there is systemic recurrence (25%) palliate with immune therapy. 

Great answers. We also have a study that allows any chemo duration for stage 3 and if ctDNA is positive, could get PD1.

Approximately 60% of the stage III dMMR colon cancer patients are cured after surgical resection and at most, only 10–15% benefit from oxaliplatin-based adjuvant treatment (André et al., PMID 26527776, Tougeron et al., PMID 26839356). There is considerable prognostic heterogeneity depending on patient and disease characteristics in this setting (Zaanan et al., PMID 28983557). Thus, the added value of adjuvant chemotherapy in the dMMR stage III colon cancer population is not very clear. Prognostication of this patient population is necessary to develop specific immunotherapeutic and/or immuno-chemotherapeutic strategies. While the low-risk dMMR stage III colon cancer patients have a 90% three-year survival, this number goes down to 60–65% for the high-risk group, i.e.T4 N2, stage III dMMR colon cancer population (Sinicrope et al., PMID 24019539).

Therefore, depending on the risk group (high or low), immune clinical trials of checkpoint inhibitors versus chemotherapy for low-risk patients, or clinical trials of chemotherapy with and without immune checkpoint inhibitors for high-risk patients would be most appropriate. Currently, available clinical trials in this setting include two phase III randomized trials testing immune chemotherapeutic strategies: the ATOMIC trial (FOLFOX +/− atezolizumab), and the POLEM trial (24 weeks of single agent fluoropyrimidine chemotherapy or 12 weeks of doublet, oxaliplatin-based chemotherapy +/− avelumab) (Lau et al., PMID 32079623). Also, there is an ongoing randomized double-blind phase II study of adjuvant pembrolizumab versus placebo in patients with dMMR solid tumors with persistent circulating tumor DNA following surgery. (NCT03832569).

For the patient in question, first, I would look at the T and N stage and determine if this patient is in low-risk or high-risk stage III colon cancer. Then, if possible, I would enroll in one of the above-mentioned clinical trials. If a clinical trial is not possible (unfortunately more likely this will be the case due to inadequate awareness of and participation in cancer clinical trials particularly in the community oncology setting due to regulatory barriers and more [Oncology February 2019, Vol 33 No 2]), I would treat with 3 to 6 months of fluoropyrimidine and oxaliplatin-based chemotherapy.

Finally, clinical trials of neoadjuvant chemotherapy and chemo-immunotherapy for colon cancer to date have been very promising with down-staging of disease and higher rates of complete pathological response and R0 resection rates. As data continue to unfold, neoadjuvant chemotherapy and/or immunotherapy will likely find its place in the treatment of locally advanced dMMR and pMMR colon cancer in the neoadjuvant setting (Oncology, February 2022 Vol 36, Issue 2).

Updated answer: If the patient is fit enough to tolerate oxaliplatin-based chemotherapy (based on geriatric assessment, functional status, and comorbidities), FOLFOX or CAPEOX + atezolizumab is the preferred regimen, as the ATOMIC trial demonstrated benefit across age groups. If low-risk stage III, 3 months of CAPEOX + atezolizumab would minimize oxaliplatin exposure while maintaining efficacy. Whether atezolizumab alone (without chemotherapy) would be beneficial in the adjuvant setting for dMMR tumors is not established, as the ATOMIC trial only tested atezolizumab combined with mFOLFOX6, and exploratory data suggested that adequate chemotherapy exposure (>6 cycles) may be critical for realizing the immunotherapy benefit.