For decades, resection of the primary tumor, either up front or after neoadjuvant therapy, has remained the standard of care for patients with early-stage (stage I-III) colorectal cancer. Recently, a plethora of data have been published that might change the current surgery-centered paradigm.
For decades, resection of the primary tumor, either up front or after neoadjuvant therapy, has remained the standard of care (SOC) for patients with early-stage (stage I-III) colorectal cancer (CRC).1 Recently, a plethora of data have been published that might change the current surgery-centered paradigm.
A subset of patients with CRC harbors a unique genomic construct in the tumors, referred to as deficient mismatch repair(dMMR)/microsatellite instability–high (MSI-H) tumors, that leads to a high tumor mutational burden and predicts responsiveness to immune checkpoint inhibitors (ICI).2,3 Several prospective studies have demonstrated remarkable antitumor activity with ICI-based immunotherapy in patients with advanced CRC.2-5 Decades of research have taught us that systemic therapy with remarkable antitumor activity in the advanced setting improves survival when used as neoadjuvant therapy––a principle firmly established in the breast cancer literature, especially with HER2-directed therapy. As a result, there is a growing interest in investigating ICIs in patients with early-stage dMMR CRC.
Several strong arguments favor exploring preoperative or neoadjuvant immunotherapy in patients with dMMR early-stage CRC. First, patients with early-stage disease are likely to respond better to neoadjuvant immunotherapy than those with advanced-stage disease owing to the higher T-cell infiltration in the tumor, lesser degree of systemic immunosuppression, and smaller tumor burden with a lower degree of clonal heterogeneity in the cancer cell population.6 Second, preclinical animal models have demonstrated that immunotherapy in the neoadjuvant setting leads to superior tumor-specific T-cell response and better antitumor activity, as stimulating the immune system with ICIs in the presence of tumor enables tumor antigen–driven T-cell expansion, which is markedly reduced if the tumor is removed.7 Third, neoadjuvant immunotherapy often shrinks the tumor, enabling margin-negative tumor resection in a higher proportion of patients and provides an opportunity to assess pathologic response that has well-known therapeutic and prognostic implications. Finally, the current SOC for early-stage dMMR CRC is associated with cancer recurrence in up to 40% of patients,8,9 likely owing to the poor antitumor activity of chemotherapy against dMMR tumors.10,11 Hence, neoadjuvant immunotherapy in early-stage dMMR CRC is an area of intense clinical investigation.
Prospective studies that fueled growing enthusiasm in neoadjuvant immunotherapy include NICHE-1 (NCT03026140),6 NICHE-2 [EudraCT 016-002940-17],12 PICC, (NCT03926338)13, and a study by Cercek and colleagues14 in locally advanced dMMR rectal cancer. Several other single-center studies and real-world series have also provided robust data to support the neoadjuvant immunotherapy paradigm.15-17 In NICHE-2,12 the pathologic complete response (pCR) rate among 107 patients with localized dMMR colon cancer receiving neoadjuvant immunotherapy with a dual ICI combination (1 dose of ipilimumab [Yervoy] and 2 doses of nivolumab [Opdivo]) was 67%. In this study, 100% of patients underwent R0 resection, and surgery was delayed only in 2 patients because of neoadjuvant immunotherapy–related toxicity. The study by Cercek and colleagues reported clinical complete response (cCR) in 12 patients with localized dMMR rectal cancer receiving 6 months of neoadjuvant immunotherapy with a PD-1 antagonist, dostarlimab-gxly (Jemperli).14
Recently, a flurry of data have been published, further supporting the role of neoadjuvant immunotherapy in patients with localized dMMR CRC. A phase 2 study with 35 early-stage patients with dMMR solid tumors receiving pembrolizumab (Keytruda)-based neoadjuvant immunotherapy reported a 65% pCR rate among the patients who underwent surgery (n = 17).18 A systematic review of published studies by our group reported a 69% pCR rate among the patients who underwent neoadjuvant immunotherapy followed by surgery.19 It is important to highlight that cancer progression or death during neoadjuvant immunotherapy was not reported in any of the studies, underscoring the safety of this strategy. None of the patients included in the systemic review had cancer relapse, although the follow-up is short. An extraordinarily high pCR rate with neoadjuvant immunotherapy predicts favorable long-term outcomes, given the consistently demonstrated strong relationship between pCR and long-term survival in many studies with neoadjuvant immunotherapy.20,21
The extraordinary pCR rate achieved with neoadjuvant immunotherapy in the subgroup of patients with early-stage CRC with dMMR tumors promises a possibility of an organ-preservation strategy. Although the organ-preservation strategy is not the current SOC in any subset of patients with colon cancer, surgically unfit patients with dMMR tumors often receive immunotherapy only and frequently achieve complete remission. Our group presented data on 24 surgically unfit patients with dMMR CRC who received neoadjuvant immunotherapy in a real-world setting. This study showed an overall response rate (ORR) of 74% with a 57% CR (pCR+cCR) rate.22 In this group, 79% of patients showed no sign of progression after a median follow-up of 12.5 months (range, 2-69). In the phase 2 study mentioned above,18 18 patients did not pursue surgery after neoadjuvant immunotherapy. After a median follow-up of 38 weeks (range, 0-103) from the last dose of pembrolizumab, 14 of 18 patients remained progression free.
These data pave the way toward a nonoperative organ-sparing approach, although several critical barriers must be overcome. The most critical barrier is the lack of a reliable tool to assess tumor response to neoadjuvant immunotherapy. It would be essential to recognize the disease progression during treatment with neoadjuvant immunotherapy to avoid missing the window of curative resection, although disease progression during treatment is exceedingly rare. Circulating tumor DNA (ctDNA) could be an important tool for response assessment once validation studies confirm the accuracy of this approach. The other barriers include the unknown optimal duration of neoadjuvant immunotherapy and a lack of consensus with respect to the optimal neoadjuvant immunotherapy regimen. Well-designed clinical trials will be needed to address these questions.