AZ aims at Pfizer, GSK with potential new PARP cancer drug

AstraZeneca’s Lynparza (olaparib) blazed a trail for the new class of PARP drugs when it was approved in 2014 to treat ovarian cancer. 

The drug has gone on to become a huge success, generating sales of nearly $1.8 billion in 2020 alone as part of a partnership with Merck & Co after expanded use in BRCA-mutated ovarian cancer, plus prostate and breast cancer indications.

Clovis Oncology, Pfizer, and GlaxoSmithKline have also marketed rival PARP drugs, with the latter two making significant acquisitions – Medivation and Tesaro, respectively – to add the medicines to their portfolios.

But AZ thinks it can improve on Lynparza, which works by inhibiting poly (ADP-ribose) polymerase (PARP), part of the specialist DNA damage repair system that cancer cells with BRCA and similar mutations rely on to survive.

These cancer cells have switched to this backup system to repair DNA damage and by turning it off, PARP inhibitors cause them to die from a process known as “synthetic lethality” as mutations in their genome run out of control.

AZ has been presenting preclinical data from AZD5305, its next-generation PARP1 inhibitor at the 2021 American Association for Cancer Research meeting, which the company thinks could be more effective and have fewer side effects compared with the first generation from this class.

In an interview with pharmaphorum, Mark O’Connor, chief scientist of early oncology discovery at AstraZeneca, said the company had decided to focus on inhibiting PARP1, one of the three main proteins from the family.

AZ’s research showed that inhibiting PARP3 doesn’t add to efficacy and inhibiting PARP2 can have a knock-on effect by interfering with production of blood cells elsewhere in the body.

O’Connor said the company has been able to design an inhibitor that does not work on PARP2 and PARP3, while being much more effective at “trapping” PARP1 to inhibit the DNA damage repair mechanism.

“There was a real reason to think that dialling out PARP2 might make PARP inhibitors better tolerated both as monotherapy and in combination with chemotherapy,” he said.

All of the approved drugs in the class are hitting PARP2 and this could give AZD5305 an advantage, said O’Connor.

He said: “That would give you potentially the same level of activity and better tolerability.”

“If you add on the fact we have improved the PARP trapping compared with the majority of inhibitors and we have maintained that level of selectivity…all of that is making us excited about the fact we could improve on the PARP inhibitors out there.

“We could have an agent that is potentially more effective as a monotherapy, better tolerated as a long-term treatment, which we know is going to be important.

“And also what we are going to be testing in the clinic is whether we can combine more effectively with standard of care chemotherapy, which would really broaden the use of PARP inhibitors.”

The phase 1 PETRA trial of AZD5305 has already begun in solid tumours and O’Connor said the drug will likely be used in cancers with BRCA mutation and could be more effective in other BRCA-like mutations where this drug class has shown activity.

Because it’s less likely to be toxic there is also the possibility of using it in combination with the chemotherapy temozolamide, a combination that has been tried in the past but proved to be too toxic with the first drugs from the class.

“We could expand significantly beyond the tumours that are being treated at the moment.”

One approach that AZ won’t be trying is marketing the drug in all forms of solid tumours – there will be many cancers that won’t respond to this type of therapy and trying this approach will dilute their effect in a trial, O’Connor said.

“I know that some companies are kind of pushing this idea that it just works in all comers but I don’t think that’s following the science,” he said.

O’Connor didn’t want to comment on whether AZD5305 could lead to an extension of the partnership with Merck & Co.

“We are looking at the early clinical data that we get with 5305 and see what that looks like,” he said.