OncoPrism-NSCLC Test Predicts Key Clinical Outcomes in Lung Cancer

Results from the observational PREDAPT trial (NCT04510129) supported the utility of the OncoPrism-non–small cell lung cancer (NSCLC) test, a clinical tool to identify which patients with late-stage NSCLC are most likely to benefit from immune checkpoint inhibitor (ICI) therapy, according to a press release from the developer, Cofactor Genomics.¹ A preprint of these results was published on MedRxiv.²

Among 195 patients with stage IIIB or IV NSCLC who were treated with an anti–PD-L1 ICI after 1 or 2 prior chemotherapy regimens in the phase 3 OAK trial (NCT02008227), those with a high OncoPrism measurement had a median progression-free survival (PFS) of 4.7 months, which was higher than the median PFS in both the OncoPrism medium and low groups, which were 2.8 months and 1.5 months, respectively (P <.001). Similarly, the median OS was 20.4 months in the OncoPrism high group, 10.5 months in the OncoPrism medium group, and 12.7 months in the OncoPrism low group (P = .043).

In the PD-1 inhibitor group, which included 89 samples and data generated from the PREDAPT trial, the overall response rate (ORR) trended upwards from the low to medium to high OncoPrism groups (P = .0083), with respective ORRs of 36%, 36%, and 71%. The median PFS was 7.0 months, 5.5 months, and 18.6 months in the OncoPrism low, medium, and high groups, respectively (P = .0038). The median OS was 13.0 months, 9.2 months, and 22.9 months, respectively (P = .011).

The performance of OncoPrism-NSCLC was compared with the performance of PD-L1 in PREDAPT, and results from the eligible 83 patients were utilized. The OncoPrism-NSCLC test achieved numerically higher accuracy, sensitivity, specificity, positive predictive power, and negative predictive power vs PD-L1 tumor proportion score (TPS). Notably, there was a significant association between OncoPrism group and response (P = .008), and there was a significant but modest association between OncoPrism-NSCLC and PD-L1 TPS (P = .006). PD-L1 TPS was associated with treatment type, as more monotherapy was observed among patients with a TPS of 50 or higher (P <.001). OncoPrism groups were also associated with disease histology, as those in the high group had fewer squamous cancers (P = .028).

Additionally, an analysis was run to evaluate OncoPrism in patients who were not treated with an ICI. OncoPrism did not predict PFS or OS in patients treated with docetaxel (n = 193), nor was it predictive of OS in the Cancer Genome Atlas (TCGA) Program data set, which was composed of 1010 patients treated with a variety of non-ICI therapies.

“For more than a decade, clinicians have lacked a reliable test that predicts which [patients with] lung cancer will benefit from immunotherapy—leaving physicians and patients to make high-stakes decisions with limited information,” stated Jarret Glasscock, PhD, chief technology officer of Cofactor Genomics and senior author on the publication, in the press release.¹ “This study, drawing on nearly 1500 [patient with] NSCLC samples from 4 independent cohorts, finally delivers what the field has been asking for: a powerful immunotherapy predictive diagnostic, rather than a prognostic marker with limited clinical utility. OncoPrism-NSCLC brings precision medicine to immunotherapy.”

Four datasets were included in the study. The first included patients in the TCGA dataset, which was filtered for primary tumor samples (n = 1010); the second included patients from the OAK trial who were treated with docetaxel (n = 193); the third included patients treated with atezolizumab (Tecentriq) in the OAK trial (n = 195); and the fourth included samples from the anti–PD-1 PREDAPT ICI cohort (n = 89).

Eligible patients in the trial had stage IV NSCLC at diagnosis, receipt of at least 2 doses of anti–PD-1 immunotherapy as first-line single-agent treatment or in combination with chemotherapy, and an ECOG performance status of 0 to 2 prior to initiation of anti–PD-1 therapy.

In the press release, the investigators noted that these results provide a “clinically actionable alternative to PD-L1 TPS” to oncologists treating patients with NSCLC.

References

1. Cofactor Genomics advances predictive cancer diagnostics with landmark OncoPrism-NSCLC study. News release. Cofactor Genomics. December 17, 2025. Accessed December 19, 2025. https://tinyurl.com/jj8bvwbn
2. Flanagan KC, Earls J, Hiken J, et al. Validation of a tissue-based predictive RNA test for immunotherapy benefit in NSCLC: the PREDAPT study. MedRxiv. Published online on December 15, 2025. doi:10.64898/2025.12.12.25342169