Frontline Pembrolizumab Approved in Europe for Metastatic MSI-H/dMMR CRC

The European Commission has approved single-agent pembrolizumab (Keytruda) for use in the frontline treatment of patients with metastatic microsatellite instability–high (MSI-H) or mismatch repair deficient (dMMR) colorectal cancer (CRC).¹

The decision was based on data from the pivotal phase 3 KEYNOTE-177 trial (NCT02563002), which demonstrated that the immunotherapy resulted in a significant 40% reduction in the risk of disease progression or death (HR, 0.60; 95% CI, 0.45-0.80; P = .0002) compared with investigator’s choice of oxaliplatin, leucovorin, and fluorouracil (FU; mFOLFOX6) with or without bevacizumab (Avastin) or cetuximab (Erbitux); or irinotecan, leucovorin, and FU with or without bevacizumab or cetuximab.²

The median progression-free survival (PFS) was more than doubled with pembrolizumab versus chemotherapy, at 16.5 months versus 8.2 months, respectively. Moreover, the 24-month PFS rate in the investigative arm was 48.3% versus 18.6% in the control arm.

“Before the KEYNOTE-177 trial, conventional chemotherapy with targeted therapy was the standard of care for patients with metastatic CRC who have tumors that are MSI-H/dMMR,” Thierry André, MD, professor of medical oncology at Sorbonne University and head of the medical oncology department at St. Antoine Hospital, Assistance Publique Hôpitaux de Paris, stated in a press release. “With this approval, patients with metastatic CRC that is MSI-H or dMMR status will gain a monotherapy treatment option that has shown superior PFS compared with standard-of-care chemotherapy.”

The multicenter, open-label, active-controlled, phase 3 KEYNOTE-177 trial enrolled a total of 307 treatment-naïve patients with MSI-H/dMMR stage IV CRC. To be eligible for enrollment, patients had to have an ECOG performance status of 0-1 and measurable disease per RECIST v1.1 criteria.

In the trial, patients were randomized 1:1 to receive intravenous (IV) pembrolizumab at a dose of 200 mg once every 3 weeks for up to 35 treatment cycles (n = 153) or to investigator’s choice of IV chemotherapy with mFOLFOX6 plus bevacizumab or cetuximab or FOLFIRI plus bevacizumab or cetuximab, administered once every 2 weeks (n = 154).

Notably, crossover from the control arm to the pembrolizumab arm was allowed for patients with confirmed progressive disease per RECIST v1.1 criteria and central review. Treatment was given until intolerable toxicity, progressive disease, or the patient or physician decided to withdraw from the trial.

The co-primary end points were PFS per RECIST v1.1 criteria and per blinded independent central review (BICR) and overall survival. Key secondary end points comprised objective response rate (ORR) per RECIST v1.1 criteria and BICR, as well as safety. Investigators evaluated tumor response at week 9 and every 9 weeks thereafter.

The median age of participants was 62.7 years, 49.5% were male, 52% had an ECOG performance status of 0, and 50% had metachronous disease. The majority of the participants were from Western Europe and North America and most had right-sided disease. Twenty-three percent had previously received adjuvant treatment and 4% had neoadjuvant therapy. About 23% of patients had BRAF, KRAS, or NRAS wild-type disease, 25% had BRAF V600E–mutated disease, and 24.5% had KRAS- or NRAS-mutated disease.

Additional data from the trial presented during the 2020 ASCO Virtual Scientific Program showed that the PFS benefit observed with pembrolizumab was upheld across all subgroups analyzed except for those who had KRAS- or NRAS­-mutated disease.

Pembrolizumab elicited an ORR of 43.8% (n = 67) versus 33.1% (n = 51) with chemotherapy, translating to a difference estimate of 10.7 (P = .0275). In the pembrolizumab arm, the ORR was comprised of an 11.1% complete response rate, a 32.7% partial response rate, and a 20.9% stable disease rate. The disease control rate with the immunotherapy was 64.7%, and 29.4% experienced disease progression. The median time to response with pembrolizumab was 2.2 months (range, 1.8-18.8).

Moreover, the median duration of response in the pembrolizumab arm had not yet been reached (range, 2.3+ to 41.4+) versus 10.6 months in the chemotherapy arm (range, 2.8 to 37.5+).

Ninety-seven percent of patients on the investigative arm experienced all-grade toxicities versus 99% of those on the control arm; of these effects, 80% and 99%, respectively, were determined to be grade 3 or higher in severity. No patients who received pembrolizumab died versus 1% of patients who received chemotherapy. Moreover, 10% versus 6% discontinued treatment with pembrolizumab and chemotherapy, respectively.

The most frequently reported toxicities with pembrolizumab included diarrhea (all grade, 25%; grade 3 or higher, 2%), fatigue (all grade, 21%; grade 3 or higher, 2%), and nausea (all grade, 12%). The most commonly reported adverse effects (AEs) in the chemotherapy arm comprised nausea (all grade, 55%; grade 3 or higher, 2%), diarrhea (all grade, 52%; grade 3 or higher, 10%), and fatigue (all grade, 44%; grade 3 or higher, 9%).

Immune-mediated toxicities occurred in 31% of patients on the investigative arm compared with 13% of those on the control arm; of these toxicities, 9% and 2%, respectively, were grade 3 or higher. Seven percent of patients who received pembrolizumab discontinued treatment due to these effects. The most common immune-related AEs on the pembrolizumab arm included hypothyroidism (all grade, 12%), colitis (all grade, 7%; grade 3 or higher, 3%), hyperthyroidism (all grade, 4%) and pneumonitis (all grade, 4%).

Thirty-six percent of patients on the control arm ended up crossing over to the pembrolizumab arm following progressive disease.

References

1. European Commission approves KEYTRUDA (pembrolizumab) as first-line treatment in adult patients with metastatic microsatellite instability-high (MSI-H) or mismatch repair deficient (dMMR) colorectal cancer. News release. Merck. January 26, 2021. Accessed January 26, 2021. http://bwnews.pr/39hrv34.
2. Andre T. Shiu Kai-Keen, Kim TW, et al. Pembrolizumab versus chemotherapy for microsatellite instability-high/mismatch repair deficient metastatic colorectal cancer: The phase 3 KEYNOTE-177 study. J Clin Oncol. 2020;38(suppl 15):LBA4. doi:10.1200/JCO.2020.38.15_suppl.LBA4