Identifying Patients With Pancreatic Cancer Likely to Respond

Pam Harrison

June 02, 2020

The presence of either germline or somatic mutations that encode proteins involved in a form of DNA repair known as homologous recombination (HR) identifies patients with advanced pancreatic cancer who are going to respond best to first-line (1L) platinum therapy, new research shows.

From 5% to 9% of patients with pancreatic cancer harbor germline or somatic mutations in the core HR genes (BRCA1, BRCA2, PALB2).

As well as identifying patients most likely to respond to therapy, "the identification of a validated biomarker for treatment selection may also reduce the likelihood of a patient receiving an ineffective therapy," senior author Eileen O'Reilly, MD, Memorial Sloan Kettering Cancer Center, New York City, told Medscape Medical News.

"The current treatment paradigm for patients with a germline BRCA mutation is platinum therapy followed by consideration of PARP inhibitor maintenance," she explained.

"Our study addresses genes beyond BRCA and includes somatic mutations as well, and [we were] aiming to understand the allelic status of homologous recombination deficiency (HRD) genes and how this may predict for benefit to platinum agents, PARP inhibitors, and possibly immunotherapy," she added.

"Any ability to refine a treatment approach for a given patient is value added," she said.

The study was published online May 22 in Clinical Cancer Research.


The study involved 262 patients with metastatic pancreatic ductal adenocarcinoma (PDAC). Most patients had de novo stage IV pancreatic cancer, the authors note. First-line platinum therapy was administered to 61%.

All patients underwent germline and somatic sequencing using the MSK-IMPACT sequencing test, which employs next-generation sequencing.

In this study, the team analyzed 17 homologous recombination genes included in the germline and somatic panels of the MSK-IMPACT gene set.

Core HR genes include BRCA1, BRCA2, and PALB2; noncore HR genes include a number of others, such as AATM, CHEK2, BAP1, RAD51, and FANCA. The researchers also analyzed whether HR mutations (HRm) were present in one or both copies of each gene.

Some 19% of the group had HRm. Of these patients, 1 5% had germline mutations, and 4% had somatic mutations.

There were no differences in genomic instability or outcomes between patients with germline mutations and those with somatic mutations, and so the two groups were analyzed together. These patients were considered to have homologous recombination deficiency, or HRD.

At a median follow-up among patients who survived for 21.9 months, "the median OS [overall survival] of the entire cohort was 15.5 months," the authors report.

However, for HRD patients who underwent treatment with 1L platinum therapy, median OS was 25.1. For non-HRD patients who underwent treatment with or without 1L platinum therapy, the median OS was 15.3 months.

On multivariate analysis, for patients with HRD, risk for all-cause mortality was reduced by 50% in comparison with patients who did not have HRD (hazard ratio [HR], 0.50; P < .01).

"The median PFS (mPFS) of the entire cohort was 7.1.months," the investigators report.

Again, however, for patients with HRD who underwent treatment with 1L platinum, mPFS was significantly longer in comparison with patients who did not undergo treatment with 1L platinum (12.6 vs 4.4 months).

Importantly as well, among patients with mutations in both gene copies, PFS was longer for those who received platinum-based chemotherapy, at 13.3 months, than for patients treated with other therapies, at only 3.8 months (P < .0001).

The same effect was not seen among patients with mutations in only one gene copy, clearly suggesting that platinum-based chemotherapy is likely to lead to greater clinical benefit for patients with mutations in both copies of HR genes, the researchers comment.

Furthermore, PFS was approximately 56% both for patients with core HRm and those with noncore HRm who underwent treatment with 1L platinum therapy, they add.

O'Reilly emphasized that these findings do not mean that patients with PDAC who do not have HRD should not receive treatment with either gemcitabine plus nab-paclitaxel or FOLFIRINOX, both of which are standard treatments for metastatic PDAC.

"What we can say at the current time is that having biallelic loss in a core gene — BRCA1 or 2 — maximizes the chances of responding to platinum and potentially to a PARP inhibitor," she said.

A patient with a monoallelic loss in an HRD gene may still benefit from the same treatment strategy, she commented. "However, it is likely that the benefit will be less compared to patients with biallelic loss."

Waiting for Test Results

The authors did point out that it can take 3 to 6 weeks for results of tissue-based genomic sequencing to be available for use in making first-line treatment decisions.

This may limit the benefit of genomic sequencing in a patient population whose median OS is less than 1 year.

On the other hand, germline test results take less time time to arrive, so it is more feasible to wait for such results, O'Reilly pointed out.

She also noted that so-called liquid biopsy is improving, and as the sensitivity gets better, that approach may provide timely results for clinical decision making.

O'Reilly is a paid consultant for Ipsen, Merck, Sobi, Cytomx Therapeutics, Celgene, and Polaris. Rafael has received commercial research grants from Genentech, BMS, Celgene, MabVax Therapeutics, AstraZeneca, Silenseed, and ActaBiologica.

Clin Cancer Res. Published online May 22, 2020. Abstract

Follow Medscape Oncology on Twitter for more cancer news: @MedscapeOnc.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.