A whole-genome sequencing approach shows early promise over current commercial methods for identifying more patients likely to benefit from PARP inhibitor cancer treatments, according to a study led by Weill Cornell Medicine and NewYork-Presbyterian investigators. The findings suggest further development of this approach is merited.
In the study , published Jan. 12 in Communications Medicine, the researchers performed whole-genome sequencing analysis on hundreds of tumor samples obtained by informed consent as part of a precision medicine initiative by Weill Cornell, NewYork-Presbyterian and Illumina, Inc. , a biotechnology company known for its DNA sequencing technology. They used the results to train and validate an algorithm that detects homologous recombination deficiency, a type of DNA-repair defect. Tumors with this defect are vulnerable to PARP inhibitors, which further disrupt DNA repair, causing cancer cells to accumulate DNA damage and die. Platinum-based chemotherapies, which also damage DNA, tend to work better in these cases. An initial test suggested that the algorithm is more accurate in predicting PARP-inhibitor treatment responses compared with existing methods.
"A comprehensive analysis of the entire genome has advantages compared with traditional, targeted detection strategies for predicting homologous recombination deficiency," said study senior author Dr. Juan Miguel Mosquera , professor of pathology and laboratory medicine and director of research pathology at the Englander Institute for Precision Medicine at Weill Cornell and a pathologist at NewYork-Presbyterian/Weill Cornell Medical Center. Dr. Mosquera is also a member of the Sandra and Edward Meyer Cancer Center at Weill Cornell.
The study was a collaboration with the medical diagnostics company Isabl, Inc .
Until now, clinicians have focused primarily on BRCA1 and BRCA2 mutations, the most frequent drivers of this DNA-repair defect, to determine if patients will benefit from PARP inhibitors. These mutations can be found most frequently in patients with breast, ovarian, pancreatic and prostate cancers. However, research shows that many other gene mutations can also disrupt this repair process—and in recent years, whole-genome sequencing, which can detect these broader changes, has become affordable enough for routine use.
The team used 305 samples from Weill Cornell and NewYork-Presbyterian patients with various cancers to train an algorithm that was developed by Isabl. The algorithm looks for a genome-wide variety of DNA defects that are known to be associated with homologous recombination repair deficiency. They then validated the algorithm using a cohort of 556 cancers and tested it against commercial methods using an additional 212 tumor samples.
The algorithm detected the DNA-repair deficiency in many of the samples, including 21% of breast tumors, 20% of pancreatic and bile duct tumors, and 17% of gynecological tumors. Notably, 24% of the detected cases did not involve BRCA1 or BRCA2 mutations, highlighting the diversity of the underlying genetic mutations.
In several cases, the algorithm appeared to flag "false negative" and "false positive" predictions from the commercial method that didn't match patient outcomes.
The team plans to conduct larger studies of the new detection algorithm as a general tool to guide cancer treatment.
