Expression of programmed death-1 (PD1) mRNA may predict outcomes after anti-PD1 therapy across cancer types, according to investigators.
High levels of PD1 mRNA were significantly associated with response to anti-PD1 monotherapy, investigators found in an analysis of tumor samples from 117 patients with advanced cancers who had received either nivolumab or pembrolizumab.
Further validation of PD1 mRNA is warranted to help select patients who might benefit from an anti-PD1 treatment strategy, wrote investigator Aleix Prat, MD, PhD, of Hospital Clínic of Barcelona, and his coinvestigators.
“Identification of reproducible biomarkers that can be applied to predict benefit of anti-PD1 monotherapy might be of clinical value,” Dr. Prat and his coinvestigators note. The report is in Annals of Oncology.
Previous studies support use of PDL1 expression by immunohistochemistry as a biomarker for pembrolizumab in non–small-cell lung cancer; however, that biomarker has some technical limitations, and has not been predictive in other cancer types and with other anti-PD1 drugs including nivolumab, Dr. Prat and his coinvestigators said.
The 117 tumor samples evaluated for PD1 mRNA expression comprised 59 advanced melanomas, 32 non–small-cell lung cancers, 14 renal cell cancers, and 12 other tumors, according to the report. Sixty-two of the patients had been treated with pembrolizumab, and 55 received nivolumab.
About one-quarter of the samples (28.2%) were classified as “PD1-high” with a preestablished cutoff value developed by Dr. Prat and his coinvestigators.
The overall response rate was 51.5% for the patients who had PD1-high tumors, versus 23.8% for the remaining tumors (P less than .001). Those non-PD1-high tumors, when grouped as PD1-intermediate and PD1-low, had overall response rates of 26.6% and 15.0%, respectively.
Median progression-free survival was 8.17 months for PD1-high tumors and 3.18 months for the rest of the tumors (P = .011), the report shows. Similarly, overall survival was a median of 23.4 months for PD1-high tumors and 14.9 months for the rest (P = .330).
Dr. Prat and his colleagues detailed earlier investigations validating PD1 mRNA as a biomarker, including an analysis of PD1 and immune-related gene expression in 10,078 samples from 34 cancer types in The Cancer Genome Atlas.
In that analysis, PD1 was strongly correlated with a group of 30 genes that were “significantly enriched” in biological processes including CD8-T-cell activation, the investigators said.
Moreover, high levels of PD1 mRNA expression were strongly correlated with overall response rates reported in the literature for anti-PD1 monotherapy, they added.
They also reported results of an analysis they used to develop the PD1-high cutoff value. That analysis was based on PD1 mRNA expression in 773 tumor samples across 17 tumor types.
“Our results are consistent with the hypothesis that identification of a preexisting and stable adaptive immune response using PD1 mRNA expression predicts outcome across cancer-types following anti-PD1 monotherapy,” the researchers wrote.
The work was partially sponsored by Instituto de Salud Carlos III, Spanish Society of Medical Oncology, Banco Bilbao Vizcaya Argentaria Foundation, Pas a Pas, Save the Mama, and the Breast Cancer Research Foundation. Dr. Prat disclosed an advisory role with Nanostring Technologies.
SOURCE: Paré L et al. Ann Oncol. 2018 Aug 27. doi: 10.1093/annonc/mdy335.