NEW YORK (GenomeWeb) – New data published this week provides a foundation for future tests to predict prostate cancer patients' response to PARP inhibitors, with added potential for monitoring emerging resistance so that treatment can be altered in a more dynamic and personalized manner.
The results, published this week in Cancer Discovery by a team from the UK's Institute for Cancer Research, reflect a burgeoning effort to lay the groundwork for the validation of liquid biopsy tests that not only detect the molecular features of a cancer, but also track its response to treatment and its evolution in response to specific drugs.
This growing arm of research reflects strong interest by physicians in tools that give them more precise and immediate insight into whether a patient is on the best drug, given the increasingly complex landscape of therapies.
ICR researcher Joaquin Mateo, a first author of the study, said that clinicians who work with prostate cancer patients are particularly anxious for more dynamic tools because existing technologies, like CT-scans, don't offer much in the way of dynamic or real-time assessment of disease progression or drug response.
"A big proportion of patients only have lesions in the bones, so we are a bit blinded sometimes for making decisions," Mateo said. "That creates a lot of interest in biomarkers to guide decisions."
Although some media reports this week have highlighted the team's publication as heralding an exciting new prostate cancer test, Mateo clarified that the work so far has been preliminary, and that he and his colleagues don't have plans to develop the test commercially.
The investigators also haven't developed a new test per se. In fact, the methods they used in the study were standard – targeted next-gen sequencing, some exome-sequencing, and simple quantification of levels of circulating cell-free DNA.
That said, Mateo and colleagues highlighted significant practical advantages that this type of testing approach would offer if the team's initial results can be further validated. This would include the ability to non-invasively test patients for eligibility for PARP inhibiting drugs, to identify early signs of treatment failure or resistance, and potentially to detect specific resistance mechanisms that would impact subsequent therapy choices.
In the study, the researchers analyzed blood serum samples from a small cohort of advanced prostate cancer patients treated with AstraZeneca's Lynparza (olaparib) in the TOPARP-A trial. Serial blood samples were available for 46 of the trial's 49 patients.
Comparing their results to known patient outcomes, the researchers were able to pick out which prostate cancer sufferers responded to PARP inhibition in the trial in the presence of blood-borne mutations, like BRCA1/2, that are associated with homologous recombination deficiency (HRD).
Alongside tests for BRCA mutations, the broader HRD category has been the subject of increasing attention from diagnostics developers and pharmaceutical companies developing PARP inhibitors.
But so far, it's still unclear how the landscape will resolve in terms of approved tests and their indication as either companion diagnostics, which are required for the prescription of a drug, or complementary diagnostics, which leave decision-making up to clinicians.
Investigation of these drugs in prostate cancer is still relatively new, but in ovarian cancer, for example, the US Food and Drug Administration recently approved a test from Foundation Medicine, FoundationFocus CDxBRCA, as a companion diagnostic to Clovis Oncology's PARP inhibitor Rubraca (rucaparib). At the same time, Clovis and Foundation have also said that they are developing a HRD signature that they hope can identify a greater proportion of patients who will respond to Rubraca.
The FDA also approved Myriad Genetics' BRACAnalysis CDx as a companion diagnostic forolaparib in ovarian cancer, but approved the same test as complementary for Tesaro's Zejula (niraparib).
Myriad has also submitted an HRD test for approval alongside niraparib, although the future of that test is unclear, after Tesaro recently said that it was now planning to pursue approval of the drug in patients regardless of either BRCA or HRD biomarkers.
Mateo and other researchers involved in the new liquid biopsy study had previously reported that up to 30 percent of lethal prostate cancers were characterized by alterations in genes linked to HRD, including BRCA2, ATM, BRCA1, PALB2, FANCA, CHEK2, and CDK12. Olaparib was subsequently granted breakthrough designation by the FDA for such patients, based on data from the TOPARP-A trial.
In the new study, the investigators tested blood samples from six TOPARP-A subjects that had these HRD-associated somatic mutations in their tumor tissue. All six showed the same mutations in their baseline circulating cell-free DNA.
Though very preliminary, the fact that it was possible to pick up these alterations in the plasma of all the patients who showed them in their tumor tissue suggests that the future of HRD testing to guide PARP inhibitors could be blood based, at least for prostate cancer patients, the authors wrote.
"Our trial published just a year and a half ago on these same patients was the first ever to report that [these] drugs work in this disease," Mateo said. "And now there are so many efforts to try to optimize what is the best biomarker, whether this can be done from plasma or you need tissue, and whether you need to look at specific mutations or can derive signatures [for HRD]."
"There are several trials now running for different PARP inhibitors, and each is looking at different strategies for selecting patients, so we will really need studies comparing these strategies to know what is best," he added. "But plasma has the advantage that it is easy to track, so you can detect easily these changes that are definitely not going to be captured by indirect signatures and can be difficult to obtain from biopsies."
This potential of ctDNA to help doctors decide not only which patients to treat with these drugs, but also to identify emerging resistance or failure to respond and quickly change treatment, was also borne out in the study this week.
Investigators were able, for example, to see signs of non-response to olaparib in blood samples taken after therapy, based on the overall quantity of circulating cancer DNA. And when they looked at the DNA of patients when they later showed signs of relapse, they could detect specific mutational features linked to treatment resistance.
In the five HRD patients that responded to therapy, for example, the allele frequencies of somatic mutations all decreased to less than 5 percent following olaparib treatment. In the the sixth, nonresponding patient, on the other hand, mutation levels were unchanged throughout therapy.
Over the full study population, the investigators calculated that patients who showed more than a 50 percent decline in overall cfDNA after four weeks had a significantly longer progression-free survival. A 50 percent fall in cfDNA after 8 weeks was also independently associated with longer overall survival.
Finally, investigators could also identify specific new mutations emerging at disease progression that likely represent tumor subclones induced by therapy and driving drug resistance.
Despite the fact that the results are so preliminary, Mateo and his co-authors wrote that the ability to identify emerging subclonal mutations would be of "major clinical relevance for subsequent treatment strategies for this subset of prostate cancers, and probably also for other cancer types," if it can be validated further.
"In the field of prostate cancer, we are developing PARP inhibitors, but there is also a big wave of excitement about the activity of platinum [chemotherapy], for which the target population is the same," Mateo said.
"If we can identify patients who develop resistance by restoring function of the gene – theoretically you would also be selecting patients who should not receive platinum at that point and should think of an alternate strategy," he explained.
The current study only looked at the emergence of specific resistance clones at the time of therapy failure, but Mateo said the investigators are now doing a second study in the TOPARP-2 trial, in which they are evaluating blood samples taken at multiple timepoints, with the hope that they can also pick out specific resistance markers as they emerge earlier in therapy.
Meanwhile, the ability to use simple, and inexpensive, quantitation of overall cfDNA – or of tumor mutation allele frequencies – to identify patients who are not responding early in treatment would offer doctors the ability to quickly stop what would be a futile course of therapy, and funnel patients toward other options, the authors argued.