One of the highlights of the 2019 annual meeting of the American Society of Clinical Oncology (ASCO) earlier this month were results from the phase III POLO trial, demonstrating that treatment with Lynparza (olaparib, Astrazeneca plc/Merck & Co. Inc.) after platinum chemotherapy nearly doubled the progression-free interval (progression-free survival, PFS) in a group of 154 metastatic pancreatic cancer patients with germline BRCA mutations, from 3.8 months to 7.4 months.

Also roughly doubled was the proportion of patients who had not progressed after two years, from 9.6% to 22.1%.

Overall survival (OS), remarkably, is not yet mature.

Study lead and University of Chicago professor of medicine Hedy Kindler called the results "truly remarkable for metastatic pancreatic cancer" and said the study "should lead to a new standard of care" for metastatic pancreatic cancer patients with BRCA mutations.

With the POLO trial, pancreatic cancer joins the list of tumor types where BRCA mutations confer sensitivity to PARP inhibitors. In a press release, Astrazeneca stated that it plans to "discuss these results with global health authorities as soon as possible."

The same sensitivity has been demonstrated in patients with breast, ovarian and prostate cancers. How broadly PARP inhibition might be useful in different anatomical sites remains to be fully explored. Olivier Nataf told BioWorld Insight that "we have tested the activity in many types of tumors" and have also seen signs of activity in several other tumor types, including gastric, colorectal and non-small-cell lung tumors. Nataf is vice president, U.S. oncology, at Astrazeneca.

The greatest success stories have been in tumors that are platinum-sensitive, possibly providing a biomarker for patients likely to benefit.

Beyond POLO, several hundred posters and oral sessions at the ASCO meeting are attesting to the fact that PARP inhibition, with four approved agents and at least 10 more in clinical trials, has become a major success story of targeted therapy.

Many of those presentations were about combination treatments, and during an education session at the meeting, Timothy Yap, of the MD Anderson Cancer Center, gave a broad overview of combinatorial categories for the class, as well as predictions for how those categories are likely to shift in the next few years.

One combination that has yielded an unexpected amount of benefit is the combination of PARP inhibitors and VEGF inhibitors. In a combination trial of experimental VEGF blocker cediranib (Astrazeneca plc) and Lynparza, the combination significantly improved PFS compared to Lynparza alone. Surprisingly, that benefit appeared to be driven by a benefit for patients without BRCA mutations – a subgroup that also showed a significant benefit in overall survival in the most recent interim analysis, which was published in February.

In a paper published in the May 15, 2019, issue of Science Translational Medicine, researchers from Yale University reported that cediranib suppressed the expression of BRCA1 and BRCA2 as well as another DNA repair gene, RAD51. That suppression occurred partly as a result of the hypoxia induced by angiogenesis inhibition, but also because the drug directly affected the signaling of platelet-derived growth factor (PDGFR).

In doing so, cediranib induced a form of what has been termed "chemical BRCAness" – depriving cells of working BRCA. Furthermore, the researchers showed that this effect occurred in tumor cells but not in bone marrow cells, avoiding the synergistic myeloid toxicity that has been an issue for DNA damage repair targeting. The results suggest that at least in some cases, synthetic lethality could be used even against tumor types that lack a BRCA mutation.

Another class of combination trials where Yap predicted growth was combination trials with checkpoint inhibitors. Five phase III trials combining PARP and PD-1/PD-L1 inhibition are currently ongoing.

Yap dryly noted that this is in part because combination trials with Keytruda (pembrolizumab, Merck & Co. Inc.) and its ilk have become "the chili of oncology – everyone thinks chili makes everything hotter."

But there are also reasons to think that damaging DNA repair, which increases the mutational load of tumors, could make them more sensitive to immune-oncology approaches, and that such immune stimulation may be behind some of the long-term responses seen in the SOLO-1 trial data presented at the European Society of Medical Oncology (ESMO) last fall. (See BioWorld, Oct. 23, 2018.)

Beyond PARP

The importance of DNA repair and the success of PARP inhibitors are part of what has led to a broad range of DNA damage response (DDR) inhibitors being tested in clinical trials.

So far, though, drugs targeting other aspects of the DNA damage response beyond PARP have underperformed their -parib peers.

Broadly, DNA-damaging drugs fall into two categories.

There are those, including PARP inhibitors, that directly target DNA repair proteins such as ATM and ATR.

Others attempt to push cancer cells through the cell cycle without a chance to repair their DNA, sending them over the edge. "If you can take a cancer cell and push it into mitosis, you are much more likely to have cell death," Nilofer Azad, associate professor of oncology at Johns Hopkins University School of Medicine, told the audience at an ASCO educational session on DNA damaging agents.

Azad said the phase II trial of prexasertib (LY-2606368, Eli Lilly and Co.), whose results were reported last year, produced "the first really promising data" for a non-PARP DDR inhibitor.

In that trial, researchers from the National Cancer Institute saw roughly one-third of trial participants with BRCA wild-type ovarian tumors respond to prexasertib, which is a Chk1 inhibitor.

Still, 15 years of trying has not yet managed to produce a non-PARP targeting agent that has progressed to phase III trials, let alone an approved one.

"ATR makes total sense, ATM makes total sense, CHK makes total sense. And yet," Iain McNeish, professor of oncology at Imperial College London and director of the Ovarian Cancer Action Research Centre, told BioWorld Insight.

There are several possibilities for why targeting the DDR beyond PARP has been a challenge that has so far stymied any development beyond phase II.

The most prosaic possibility is the same old story. There are "real issues with toxicity, and real quandaries about the best patient populations," Azad said.

Myeloid cells are particularly sensitive to DNA damage, and particularly when given in combination with other agents that also cause myeloid toxicity, dose reductions may be necessary that push the agents below their threshold for efficacy – a fate that befell, among other drugs, ATR inhibitor VX-970/M6620.

As single agents, such drugs have had, at best, modest clinical benefit, but also minimal toxicity, Azad said. But "that changes completely when we add in the carboplatinum."

Those challenges may yet be overcome, McNeish said, adding that temozolomide, which was approved for glioblastoma in 2005, was close to being shelved by the time its developers figured out a regimen that would allow it to be successful.

Encouraging insights into how sequencing might work were published in the June 10, 2019, issue of Cancer Cell, where researchers from MD Anderson Cancer Center reported that in preclinical ovarian cancer models, sequential treatment with a PARP inhibitor followed by inhibitors of Wee1 or ATR had the same therapeutic efficacy as concurrent treatment, but reduced toxicity. With sequential treatment, healthy cells were able to recover from the damage done by the first drug in time to withstand the second. Cancer cells, which have a higher ongoing level of replication stress, were not able to recover from the first drug to the same degree, making them selectively vulnerable to sequential treatment.

Other studies have shown that P53 is a promising biomarker for sensitivity to Wee1 inhibitors, and the two proteins may have a synthetic lethal relationship analogous to that of BRCA and PARP.

One reason for optimism is that even PARP inhibitors themselves were to an extent late bloomers. The first PARP inhibitor to enter the clinic, Rubraca (rucaparib), was originally meant as a chemotherapy potentiator and was stymied by – wait for it – myeloid toxicity. It was not until the identification of BRCA mutations as synthetic lethal with PARP inhibition that PARP inhibitors found their sweet spot.

It is also possible, though, that PARP inhibitors are achieving their therapeutic effects via additional mechanisms besides blocking the repair of single-stranded DNA breaks. McNeish said that there are 18 members of the PARP family, including mitochondrial PARPs, with multiple different functions.

"There may be other things going on that we are still not fathoming."