Efficacy of Immunotherapy in Microsatellite-Stable or Mismatch Repair Proficient Colorectal Cancer—Fact or Fiction?
Mené au Canada sur 180 patients atteints d’un cancer colorectal de stade avancé et réfractaire (âge médian : 65 ans), cet essai de phase II évalue l’efficacité, du point de vue de la survie globale, et la toxicité de l’ajout du trémélimumab et du durvalumab aux "meilleurs soins de support" (durée médiane de suivi : 15,2 mois)
Approximately 5 years ago, the first results of the activity of single-agent pembrolizumab, a programmed cell death-1 (PD-1) antibody, were reported in patients with mismatch-repair deficient (MMR-D) and/or microsatellite instable (MSI-H) metastatic colorectal cancers.1 Since then, several additional studies have confirmed the profound effect of either PD-1 antibodies as single agents or in combination with cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) inhibitors in various malignant diseases exhibiting the MSI-H/ MMR-D phenotype. In fact, in 2017 the FDA—in a historic decision—granted the first tissue- and tumor-type agnostic approval for pembrolizumab for MSI-H and/or MMR-D treatment-refractory cancers. The activity of immune checkpoint inhibitors in this group of cancers has been linked to the hypervariant phenotype observed in these tumors.1,2 The deficiency of the mismatch repair system will conceivably lead to an accumulation of DNA variations in the tumor genome, which by stochastic probability increases the likelihood for the expression of immunogenic neoantigens on the surface of the tumor cells that in turn can be the targets for immune response. This hypothesis is supported by the fact that other molecular alterations like POLE/POLD1 variations, which are associated with a very high tumor mutation burden (TMB) have also been shown to induce susceptibility to immunotherapic approaches.2,3 In metastatic colorectal cancer, only 4% to 5% of tumors are classified as MSI-H and/or MMR-D; the rest are microsatellite-stable (MSS) or mismatch repair proficient (MMR-P). So far, studies of immunotherapies either as single agent or in combination therapy in MSS and/or MMR-P cancers colorectal cancers have consistently generated disappointing results.4,5 It would be of tremendous clinical importance if the a priori nonimmunogenic MSS and/or MMR-P colorectal cancers could be rendered susceptible to immunotherapy approaches. The study by Chen and colleagues6 now describes the result of a randomized phase 2 trial in treatment-refractory metastatic colorectal cancer in which about 180 patients were randomized 2:1 to a combination of durvalumab, a PD-L1 antibody, and tremelimumab, a CTLA-4 antibody, or best supportive care (BSC) alone. Molecular profiling and assessment of the MSI-status and TMB was performed by using a circulating cell-free DNA (cfDNA) test that was successful in 168 patients. Eleven of the missing cfDNA results were in the BSC arm; both arms had 1 patient each classified as MSI-H per cfDNA analysis. Because the study was designed as a phase 2 trial, statistical assumptions were less stringent compared with a phase 3 setting, with a power of 80% and a 2-sided ? of 0.10.
JAMA Oncology , éditorial, 2019