Marwan G. Fakih, MD

In a groundbreaking neoadjuvant clinical trial of nivolumab and ipilimumab (NICHE-2), Chalabi reported unprecedented pathological responses in 112 patients with MMR-deficient, locally advanced colon cancer. The results may underestimate the efficacy of this regimen, as most patients had surgery around 6 weeks from the start of therapy, and a longer time to surgery may have increased the rate of complete responses.

Perhaps the most important conclusion from this study is highlighting a future role for short-course immunotherapy in the adjuvant or neoadjuvant settings in MMR-deficient tumors. Despite limiting treatment to a 6-week cycle of immunotherapy, no recurrences have been reported yet in this study (short follow-up). As we design future neoadjuvant and adjuvant trials with immunotherapy, it will be important to determine if a short course of therapy, such as explored on the NICHE-2 trial, is as effective as longer treatment durations. Shorter immunotherapy duration will have advantages for patients (less toxicity) and to the health care system (less cost). More importantly, a neoadjuvant strategy can, early on, stratify patients into responders and suboptimal responders and potentially further stratify them to appropriate treatments or observation in a postoperative setting.

The findings may also have additional implications in MMR-deficient stage IV colorectal cancer. Should those patients be treated for 1 year if a complete clinical response is achieved and for 2 years or more if residual measurable disease is noted? Given the excellent responses that are seen in patients with microsatellite instability-high (MSI-H) stage IV disease, is it also possible that these patients may not need protracted therapy? Perhaps shorter durations of treatment with ipilimumab and nivolumab can be incorporated in stage IV MMR-deficient colorectal cancer. Treatment could be stopped early in the event of circulating tumor DNA clearance and negative PET imaging, especially with the limited reliability of CT given the common residual measurable fibrotic masses after immunotherapy. Such concepts should be considered for future clinical trials.

Does NICHE-2 have immediate implications on the management of locally advanced MMR-deficient colon cancer? Largely, no, but with possible exceptions. For one, the accuracy of clinical staging for MSI-H colorectal cancer is poor. Nodal disease by imaging is often false-negative as a high rate of reactive lymph nodes is noted on CT or MRI. In addition, many bulky MMR-deficient tumors on imaging are pathologically T3N0 at the time of surgery, without neoadjuvant therapy. Therefore, there is a high likelihood that many patients were “truly” pathologically stage II at the time of enrollment while being clinically staged as stage III disease.

Stage T3N0 (stage II) MMR-deficient colon cancer has an excellent prognosis with surgery alone, with an estimated recurrence rate of approximately 7%. Exposing such patients to immunotherapy in a neoadjuvant setting would not be appropriate, even with the reported low risk of grade 3 to grade 4 toxicities (and likely a higher rate of grade 1 to grade 2 toxicities). Second, the follow-up for recurrence on NICHE-2 is short. Although I am confident that the data will stand after 3 years of follow-up, the 3-year DFS rate is crucial in confirming the benefits of a neoadjuvant strategy.

On a positive note, the findings support implementation of a neoadjuvant strategy, as per NICHE-2, in all MMR-deficient tumors where a surgical resection margin would be otherwise threatened. These may include bulky T4 tumors. Given the NICHE-2 efficacy data, no such patient should be exposed to neoadjuvant chemotherapy or chemoradiation strategy, and neoadjuvant immunotherapy should be considered a standard of care in such settings.

Finally, are we ready to extrapolate to a watchful waiting strategy in MMR-deficient colon cancer treated with a short course of nivolumab plus ipilimumab? Not yet. As stated above, many MMR-deficient colon cancers are cured with surgery alone with minimal to no postoperative quality-of-life implications and require minimal imaging on follow-up and no systemic treatment. Watchful waiting strategies would require immunotherapy treatment with a low but real risk for toxicities and will require frequent imaging and frequent endoscopies, both of which are taxing to the patient and health care system. However — and given the excellent responses noted on NICHE-2 — a definitive immunotherapy treatment strategy followed by observation may not be unreasonable for patients with a high risk for morbidity or mortality from surgery or where surgical intervention can result in major implications on quality of life (ie, pelvic exenteration). In such settings, the duration of therapy would still be debated, and given the short follow-up with NICHE-2, most would not be comfortable with a 6-week treatment strategy.

Although NICHE-2 may not have immediate clinical impact on how we treat MMR-deficient disease, this study will, no doubt, have a ripple effect on how we integrate immunotherapy in clinical trials across the board in cancer care, including MMR-deficient colorectal cancer.

Andrés Cervantes, MD

This study explores the role of a very active “short course” of preoperative immunotherapy and questions the need of surgery and postoperative chemotherapy in all patients in whom the primary tumor has disappeared. Standard of care for the treatment of localized colon cancer with deficient mismatch repair status consists in surgical resection followed by adjuvant chemotherapy, despite the fact that chemotherapy is not so active and a complete disappearance of the tumor in the surgical specimen is not observed with chemotherapy.

Chalabi and colleagues used a very innovative approach, consisting of a very short course of preoperative immunotherapy followed by surgical resection. [MMR deficiency] behaves as a strong predictor of the positive effect observed with this short course immunotherapy and determining [MMR deficiency] can be easily done by immunohistochemistry in the conventional pathology lab without the need of complex molecular testing. Finally, the minimal toxicity observed may also facilitate the implementation of this strategy, potentially sparing patients from surgery.