The success of a handful of pharmacogenetically-targeted cancer drugs, such as the BRAF inhibitor Zelboraf in melanoma or the ALK inhibitor Xalkori in non-small cell lung cancer, may have given rise to the assumption that any treatment that targets a particular pathway or mutation driving a tumor will benefit that patient.
Unfortunately, this doesn't always turn out to be the case since solid tumors are the result of an accumulation of mutations, with most tumors harboring between 5 and 15 driver mutations. And cancer patients are often on a cocktail of drugs that each leave their own footprint in the course of the disease.
Given this biological reality, in early January the National Cancer Institute began enrolling patients with a variety of rare and common solid tumors in a prospective study called the Molecular Profiling based Assignment of Cancer Therapeutics (M-PACT), which aims to gauge whether drugs that target certain mutations and pathways implicated but seldom studied in certain cancers, will benefit patients.
"Because of the success of some of the drugs like vemurafenib [Roche/Genentech's Zelboraf], which hits BRAF V600E [mutations], there is a presumption that if you find a mutation and if you have a drug that hits that mutation or that pathway that there will be a clinical benefit," Shivaani Kummar, head of NCI’s Developmental Therapeutics Clinic and the principal investigator of M-PACT, told PGx Reporter. "But practically speaking, that's not true."
NCI has developed a targeted sequencing assay on Ion Torrent's AmpliSeq platform, which researchers will use to test all 180 participants enrolled in the trial for 391 mutations in 20 genes in the AKT/PI3K, RAS/RAF/MEK, and DNA repair pathways.
Patients in M-PACT will be randomized to either receive a treatment targeted to their specific mutation or pathway, or they will receive an agent in the comparator arm not pharmacogentically determined. If patients progress on the comparator arm, they will have the option to cross over to PGx-guided treatment. "What this particular study is trying to test is does this [PGx] approach actually work? If you take somebody with a mutation in the PI3K pathway and you have a drug that hits that pathway, does it actually provide benefit to the patient?" Kummar said.
There are examples where prospective PGx studies have not borne out hypotheses drawn from retrospective analysis. For example, Clovis Oncology in November 2012 pulled the plug on CO-101 as a treatment for pancreatic cancer after a prospective, randomized trial showed that the investigational agent – a gemcitabine-lipid conjugate – provided no survival advantage compared to gemcitabine alone in patients with low hENT1 expression.
Previously, retrospective studies had suggested that pancreatic cancer patients with low hENT1 expression responded poorly to standard of care gemcitabine because the chemotherapeutic requires the protein cellular transporter hENT1 to enter tumor cells. Clovis had hypothesized that CO-101 would enter tumor cells through passive diffusion and therefore would benefit patients regardless of their hENT1 expression. But when in prospective analysis there was no survival difference between CO-101-treated, hENT1 low pancreatic cancer patients and the gemcitabine-treated patients, it baffled Clovis researchers as to why the two drugs "behaved exactly the same."
By performing M-PACT prospectively, NCI researchers are hoping to minimize any confounders in the hypotheses the trial generates. As of mid-January, study investigators have tested the first few cancer patients using the custom sequencing panel. Researchers chose to test for mutations from the literature that only had pre-clinical data and hadn't been studied much in certain types of cancer in humans. They excluded well-established genetic markers known to have a pharmacogenetic mechanism in bolstering response or causing drug resistance.
In order to participate in the study, cancer patients with various solid tumors have to be resistant to the standard of care. Additionally, if they harbor driver mutations, such as BRAF, EGFR, KRAS, or ALK, known to impart a treatment benefit with or resistance to certain targeted drugs, then those patients must have already received and progressed on those treatments. In this way, researchers are making sure not to favor any particular treatment arm in M-PACT.
"Whatever is known, we don't want to retest that hypothesis," Kummar said. "If you think about it, if you put somebody with an EGFR mutation and give them erlotinib [Genentech's Tarceva] then you are basically favoring that arm. You're going to see responses. So, whichever markers are proven to work [in terms of drug targeting], or are mechanisms of resistance, are already taken out of the picture."
Currently, M-PACT is studying AbbVie's investigational PARP inhibitor veliparib in combination with temozolomide; GlaxoSmithKline's MEK inhibitor Mekinist (trametinib), recently approved by the US Food and Drug Administration for patients with BRAF V600E-mutated melanoma; Novartis's mTOR-inhibiting kidney cancer drug Afinitor (everolimus); and AstraZeneca's Wee1 inhibitor MK-1775 plus carboplatin.
Investigators picked drugs that are known to hit the target pathways of interest. For example, veliparib and MK-1775 have been shown to inhibit DNA repair, Mekinist interrogates the RAS/RAF/MEK pathway, and Afinitor blocks the AKT/PI3K pathway.
"So, this won't be like a first-in-human study where you don't know it works," Kummar said. "[The drug] has to hit the target and it has to have a Phase II recommended dose already established."
M-PACT researchers will enroll patients with common and rare cancers, such as small cell lung cancer and head and neck cancer. They will also track patients' response in each of the treatment arms using built-in statistics, so in case patients in a particular arm are not responding to the regimen, investigators can drop the drugs or add agents to hit the same pathway or include additional agents and pathways.
If patients on a PGx-guided regimen have a particularly robust response, NCI plans to spin out a separate trial to study in depth just those patients with that particular tumor profile. NCI is hoping that this clinical trial model will spur development efforts for the pharmaceutical companies contributing their drugs to M-PACT.
"This is what drug developers want to do. They want to try to find molecular subsets for patients [so] that they can actually expedite [the trial]," Kummar said. M-PACT "is basically trying to identify the patients that are going to be benefiting from their therapy. And if they do further trials, they will be smaller trials with larger treatment effects that would mean that their drug would get on the market faster."
In discussions with pharma partners, Kummar said that the firms seem to like that M-PACT is enrolling patients with a variety of solid tumors. "They can look at the database for their patients and try to identify signals for proceeding where they could really form a niche for their drug," she said.
M-PACT, similar to I-SPY and I-SPY2, is part of NIH's ongoing efforts to inform personalized drug development strategies by working with industry to perform non-conventional trials of innovative and flexible design. These types of trials have provided drug developers with critical information about whether their agents would succeed in a late stage trial and benefit a particular patient subset.
For example, at the San Antonio Breast Cancer Symposium in December, researchers from the Phase II I-SPY2 trial reported that veliparib in combination with carboplatin and standard neoadjuvant chemotherapy would have a 90 percent rate of success if AbbVie studied the drug in a Phase III trial enrolling triple-negative breast cancer patients.
According to Kummar, currently the turnaround time for receiving results from the sequencing panel is approximately 10 days. As of mid-January, researchers had enrolled five patients into M-PACT.