Clinical trials for PARP inhibitor combination therapy don’t select for specific biomarkers, but they should, researchers say.
Six DNA biomarkers that predict who will benefit from a combination of PARP inhibitors and chemotherapy have been identified, giving patients a better prognosis and potentially sparing unnecessary treatment.
PARP inhibitors prevent cancer cells from repairing themselves by blocking the poly adenosine diphosphate-ribose polymerase enzyme. They are currently used as monotherapy to treat DNA repair deficient ovarian, breast, pancreatic and prostate cancers.
These targeted drugs have also been investigated as a combination treatment with chemotherapy in over 80 patient trials worldwide, in a variety of cancers, on the hypothesis that simultaneously attacking cancer cells with chemotherapy while weakening cancer cells’ ability to repair themselves will be more effective.
While pre-clinical trials have shown success regardless of the tumour’s DNA repair status, this has not translated into clinical trials. It turns out, having one of six biomarkers of DNA repair function (BRCA1, BRCA2, homologous recombination deficiency test, ATM, ERCC1 or SLFN11) makes all the difference.
“A lot of work is being done on this premise that you can just pop drugs together and – tada! – there’ll be a synergy between them,” Associate Professor Liz Caldon, head of the replication and genome stability lab at the Garvan Institute of Medical Research, told OR.
“Particularly for these drugs, [the idea is] that it will create an interaction that will kill cancer cells. But our research is really showing that’s actually not the case,” said the UNSW cancer genetics researcher.
Her team analysed data from nine studies and around 2500 participants on the effectiveness of PARP inhibitors in combination with chemotherapy or chemotherapy alone.
Progression free survival jumped 57% in the 1181 DNA repair deficient patients who received the combination therapy, whereas therapy type made no difference in the 1366 DNA repair competent patients.
Meanwhile, benefit or no, all trial participants faced the risk of high-grade adverse haematological effects.
“It’s really about considering the trade-off between picking a more aggressive treatment and the side effects,” said Professor Caldon.
“We’ve got to be really, really careful when we’re using these combinations of drugs that we are properly selecting the patients that are going to receive them.”
DNA repair status is rarely included in trial designs. In 82 clinical trials assessing the effectiveness of combination therapy, only 23 required patients to have tumours with altered DNA status. Many of the trials that did not select patients did not go on to publish results, Professor Caldon noted.
Among the other 59 trials not requiring deficient repair DNA, 19 performed a secondary analysis relating to DNA repair biomarkers. But as they were mainly in cancer types with fewer DNA repair defects, they may not have been able to draw any conclusions.
“This is actually very important research and potentially has very practical applications,” said Professor of Medical Oncology at Flinders University in Adelaide, Bogda Koczwara.
“If we want to benefit thousands of patients tomorrow, then we need to get our research design right today.
“Is this paper going to change my clinical practice tomorrow? I think the answer is no. But this paper would change my practices as a researcher.”
The study’s conclusion that these clinical trials should target patients with repair deficient biomarkers should not be surprising, said Professor Koczwara. Using a targeted treatment without focussing on the target was, she said, “counterintuitive” and had “very significant, undesirable consequences”.
“One is that we expose patients to toxicities of treatment, where we wouldn’t necessarily expect them to benefit from the treatment. But the second one is that because we’re diluting the pool of patients who are exposed to the drug where a proportion of the patients are unlikely to benefit, we may actually end up with results that underestimate the efficacy of the treatment. And we may dismiss the drug and never bring it to clinical application,” said Professor Koczwara.
“If we gave Herceptin to everybody, we would never have noticed that it was effective.”
Dr Sanjeev Kumar, a clinician scientist and medical oncologist at the Chris O’Brien Lifehouse in Sydney, said the pre-clinical and mechanistic rationale for combining PARP inhibitors and chemotherapy, with or without DNA repair defect biomarkers, was reasonable. But this research showed that the benefit was limited to those with a DNA repair deficiency, he told OR.
“Going forward, this highlights the clinical importance of delineating markers of DNA deficiency in higher risk populations (such as breast and ovarian cancer) and suggests that DNA repair biomarkers should be used to streamline patient populations recruited to studies combining PARP inhibitors with chemotherapy,” Dr Kumar said.
