乌鸦传媒

PARIS -- Nearly every single patient with mismatch repair-deficient (dMMR) colon cancer in the NICHE-2 trial achieved a pathologic response with just two cycles of neoadjuvant immunotherapy.

Patients in the single-arm study were treated with PD-1 plus CTLA-4 blockade -- nivolumab (Opdivo) plus ipilimumab (Yervoy) -- and all 112 patients successfully underwent surgical resection, with 98% heading to the operating room on time, meeting the study's primary safety endpoint, reported Myriam Chalabi, MD, of the Netherlands Cancer Institute in Amsterdam.

And with a median follow-up of 13.1 months, the disease-free survival (DFS) rate was 100%, she said at the European Society for Medical Oncology (ESMO) annual congress here, pointing out that the expected rate for this patient population would be in the vicinity of 15% by this point. The primary efficacy endpoint for the trial is DFS at 3 years, with success defined as a rate of 93% (data are expected next year).

But the accolades were reserved for a different endpoint and sustained applause rang out when Chalabi displayed the waterfall plot showing the depth of pathologic response with just 4 weeks of nivolumab plus ipilimumab.

Among the 107 patients evaluable for efficacy, all but one had a pathologic response, 95% had a major pathologic response (MPR), and 67% had a pathologic complete response (pCR) -- meaning no residual viable tumor in both the primary tumor bed and lymph nodes.

"As you can appreciate, the pathologic regression observed was near-complete or complete in almost all patients," she said.

"A waterfall plot or a tidal wave?" quipped Toni Choueiri, MD, of the Dana-Farber Cancer Institute in Boston, on .

By contrast, prior trials involving neoadjuvant chemotherapy have shown pathologic response rates in the range of 5% to 7% for this patient population, said Chalabi.

Approximately 10% to 15% of colon cancers are classified as dMMR, she explained, and these tumors are highly sensitive to immune checkpoint inhibitors, where multiple agents have been in the metastatic setting.

"Striking data," observed ESMO discussant James Larkin, MD, PhD, of the Institute of Cancer Research in London, who cautioned that the findings will need longer follow-up but were consistent with a recent report showing a 100% complete response rate with anti-PD-1 monotherapy in a small group of patients with dMMR rectal cancers.

Larkin noted that it was patients who drove the organ-preservation paradigm in rectal cancer, where pelvic radiotherapy and surgery can have major lifelong consequences in terms of morbidity and function.

"It may be with colon cancer that having a hemicolectomy plus/minus temporary stoma is less morbid," he said, but "I think it would be really important to get patient views on organ preservation in colon cancer as well."

Alexander Eggermont, MD, PhD, first up to the microphone during the Q&A session, stressed the potential impact of the findings.

Patients with dMMR tumors scheduled for resection "should be taken off the surgical program," said Eggermont, of Princess Maxima Center for Pediatric Oncology in the Netherlands.

"They should be sent to the medical oncologist for the first dose of ipi/nivo," he said. "We will live the day that they will not undergo surgery anymore after these schedules -- that's the next step."

NICHE-2 was a multicenter, single-arm study that enrolled 112 patients in the Netherlands with previously untreated non-metastatic dMMR colon cancer undergoing surgery. The first cycle of neoadjuvant treatment included ipilimumab (1 mg/kg) and nivolumab (3 mg/kg). The second cycle, given 2 weeks later, was limited to nivolumab alone. Median time to surgery from the first dose was 5.4 weeks.

The study built on, and includes patients from, the , which was a proof-of-concept study evaluating the strategy in 32 patients with early-stage dMMR colon cancer.

Eligibility requirements for NICHE-2 included clinical T3 and/or N-positive disease on radiologic staging, no clinical signs of obstruction, no clinical symptoms or radiologic suspicion of perforation, and no conditions requiring the use of systemic steroids or other immunosuppressive medications.

Pathologic response was defined as 50% or less residual viable tumor; MPR was defined as 10% or less residual viable tumor, and included patients with pCRs in the primary tumor but viable tumor in the lymph nodes.

Patients had a median age of 60 years (range 20-82), and 58% were women. About three-fourths had high-risk stage III disease, 13% had low-risk stage III disease, and 13% had stage I/II disease. Primary tumor location was the right colon for about two-thirds of the cohort. About half had radiologic high-risk disease (both T4 and N2), said Chalabi, and abdominal wall involvement was common.

Asked whether randomized data would be needed to make this approach standard, Chalabi noted the group with T4 tumors.

"I wouldn't want to randomize those patients," she said. "Surgeons usually would prefer to have some type of downstaging before continuing on to surgery in order to increase the chances of achieving tumor-free resection margins and also to limit the extent of surgery needed to achieve that."

The case for randomization is stronger in earlier disease, she said, but if recurrences can be prevented even in stages where recurrence is more rare, such as stage II tumors (about 10% at 3 years), "we're curing 10% more patients."

A little less than a third of patients had Lynch syndrome, and pCRs were more frequent in this subset (78% vs 58% in those with sporadic tumors).

Immune-related adverse events (AEs) were reported in 61% of patients, with 4% being grade 3/4. Two patients experienced AEs that caused delays in surgery of 2 weeks or more. Surgery-related AEs occurred in 21% of the patients, with 13% being grade 3/4. Anastomotic leakage or wound infections occurred in 5%.

When the prospect of a NICHE-3 trial came up, Chalabi said that ideally it would have been an international study validating the approach tested in this trial. However, a subsequent trial is being developed and will likely involve nivolumab plus anti-LAG-3 relatlimab, "which is a shame," she noted.

"If we do get similar responses with nivolumab and anti-LAG-3, then that may be an avenue to test organ-sparing approaches with that combination in this population," she added.

Comment