By Turna Ray
After a failed Phase III trial of Sanofi-Aventis' triple-negative breast cancer candidate iniparib, the drug developer has decided to conduct molecular analysis to try to identify best responders to the drug.
Data from a Phase III trial showed that triple-negative breast cancer patients receiving the PARP inhibitor with a gemcitabine and paclitaxel chemotherapy regimen did not experience significant improvements in overall and progression-free survival compared to patients receiving just chemo. However, a subgroup analysis conducted as part of the trial found that iniparib improved overall survival and progression-free survival in the second- and third-line settings.
Reviewing the data at last week's annual meeting of the American Society for Clinical Oncology, study leader Joyce O'Shaughnessy of Baylor Charles A. Sammons Cancer Center reflected that if she could do the trial over again, she would try to get a better understanding of the molecular heterogeneity within patients' tumors.
Women with triple-negative breast cancer have tumors that lack estrogen and progesterone receptors and don't overexpress the HER2 protein, but there are additional molecular subtypes even within this classification. If this study could be conducted again, "ideally we'd have a better handle on the heterogeneity of the cancer," O'Shaughnessy said during a session at the ASCO meeting in Chicago.
As part of the Phase III study, which enrolled 519 patients, O'Shaughnessy and colleagues used Affymetrix gene expression arrays to conduct preliminary molecular subtyping of 304 patients' FFPE tumor samples. They classified these patients as falling into six groups: basal-like, claudin-low, ERBB2, luminal A, luminal B, and normal-like.
According to O'Shaughnessy, the researchers will next try to gain a deeper understanding of the molecular subtypes at play in triple-negative breast cancer tumors. "[Biomarker] analyses are underway to evaluate iniparib benefit within the intrinsic subtypes, as well as within other molecularly defined subpopulations," she said.
Going forward, the researchers will further analyze the preliminary data they gathered on patients' cancer subtypes using the PAM50 test, an RT-PCR based assay that measures 50 classifier genes and five control genes. PAM50, which is marketed as a laboratory-developed test through ARUP Laboratories, classifies breast cancer patients into five intrinsic subtypes: luminal A, luminal B, HER2-enriched, basal-like, and normal-like.
Additionally, central pathology review of Phase III study participants' ER, PR and HER2 status at the end of the study is ongoing, O'Shaughnessy remarked.
"At the moment, we don't fully understand the mechanism [of iniparib], but what we do know is that there are biomarkers of DNA damage," Charles Hugh-Jones, VP of medical affairs at Sanofi Oncology, North America, told PGx Reporter. "We're actively taking this drug back and doing a huge amount of translational work to really understand that mechanism of action."
Sanofi has said it plans to discuss iniparib's Phase III data with the US Food and Drug Administration, as well as with European regulatory authorities. The company will continue to study iniparib in breast, non-small cell lung, and ovarian cancer.
Not Your Typical PARP Inhibitor
The news that Sanofi will conduct genetic analysis to attempt to identify best responders for iniparib comes after much speculation as to why the drug developer didn't use a pharmacogenomic strategy for iniparib earlier in its development.
Competitors Abbott and AstraZeneca are applying a PGx strategy in the development of their PARP inhibitors by using Myriad Genetics' BRCA mutation test to direct the therapy to best responders in clinical trials (PGx Reporter 06/30/2010). Abbott's PARP inhibitor veliparib is being developed in breast cancer and AstraZeneca's PARP inhibitor olaparib is being developed in ovarian cancer.
AstraZeneca and Abbott are studying their drugs in patients with BRCA mutations because published literature suggests that BRCA 1/2 mutations make cancer cells sensitive to PARP inhibition, resulting in apoptosis. The PARP1 enzyme and the BRCA gene work in concert to repair DNA damage, enabling survival of cancer tumors. In patients with BRCA mutations, however, PARP inhibitors can thwart repair of DNA lesions that would drive the cancer.
According to O'Shaughnessy's presentation at ASCO, however, iniparib's mechanism of action is unlike other "parib" drugs and it doesn't behave like a typical PARP inhibitor even though it has been characterized as one. Challenging this perception, O'Shaughnessy described the drug as "inducing cell cycle arrest in G2/M phase [and] inducing double cell DNA damage [at the] γ-H2AX foci," but she added that the drug "does not inhibit PARP 1 and 2 at physiologic drug concentrations."
Because researchers don't fully understand the mechanism of iniparib, it is currently unclear whether the drug can even be called a PARP inhibitor.
"Iniparib isn't typical of a PARP inhibitor at physiological concentrations in humans; on the other hand it does inhibit PARP activity at high levels," Hugh-Jones observed. "So, there's a degree of balance in terms of whether or not it's called a PARP inhibitor, but it's reasonable to say that in patients it doesn't possess the characteristics typical of the PARP inhibitor class."
Given this nuanced characterization of iniparib's mechanism of action, the molecularly defined subpopulation of patients most likely to benefit from iniparib may be different from that of other PARP inhibitors currently under development.
Following O'Shaughnessy's presentation, an audience member commented that based on Phase II data presented at last year's ASCO meeting, the research community still believed that iniparib fit into the PARP inhibitor class and, as such, during the recruitment of this Phase III study, there was "hysteria" among triple-negative breast cancer patients to try to get into this trial. The audience member noted that since the mechanism of iniparib is still not fully understood there still may be differential response among patients with BRCA mutations.
To this O'Shaughnessy commented that it is "not entirely" known how many patients in the Phase III study harbored BRCA 1 and BRCA 2 mutations, though the consent process for the study was modified to allow participants to disclose their BRCA mutation status, if it was known. Study participants could also consent to undergo BRCA mutation testing as part of the trial. "A minority of patients took up that offer so we have some information, but just a minority," she said without elaborating on what the initial data show.
Myriad markets the only commercially available test to assess BRCA mutation status. A company spokesperson told PGx Reporter that although Sanofi offered patients the chance to be tested for BRCA in the iniparib Phase III study, there is no formal agreement between the two companies for the development of a companion test. Myriad has had discussions with Sanofi about using BRACAnalysis to conduct a retrospective analysis of the Phase III patient population(PGx Reporter 02/02/2011), but no such deal has yet been announced.
[ pagebreak ]
Hugh-Jones couldn't provide any details as to whether iniparib could potentially be advanced in patients with BRCA mutations. "At the moment … we're looking at a wide range of options in terms of analyzing the data, and looking at platforms to understand how iniparib works best," he said.
According to O'Shaughnessy, the researchers will look at DNA repair signatures compared to mesenchymal-type signatures. "So we're going to have a good handle, I'm sure, on the biology [of patients who are] more deficient in DNA repair but that will probably need to be our surrogate for the BRCA 1/2 situation across all patients," she said.
It is currently unknown what additional molecular analysis Sanofi will conduct to assess iniparib response in patient subsets. In an article discussing iniparib's Phase II data, published in the New England Journal of Medicine in January, O'Shaughnessy and co-authors noted that in addition to BRCA mutations, "aberrations in MRE11–RAD50–NBS1, ATM, p53, and PALB2 … have also been implicated in the tumorigenesis of triple-negative breast cancer" and are worth exploring in drug development.
One of the findings from the iniparib Phase III trial was that patients who received iniparib plus chemotherapy in the second- and third-line setting did see a survival benefit compared to patients receiving just chemo. This will also be further investigated in future studies.
"Overall, looking at the data, we all have to accept that this was a negative Phase III study," Huge-Jones said. "Looking at the exploratory analysis that was done in second- and third-line patients, it was intriguing to see the hazard ratio and overall survival benefit" in those patients.
The improved efficacy in second- and third-line patients may be due to "chance," O'Shaughnessy reflected during her presentation, because "when you're looking at subsets you're looking at smaller numbers." Another hypothesis for why these patients with later-stage disease benefitted from iniparib is that first-line patients were sicker than second- and third-line patients. "Because of the excitement around this trial, the first-line patients entering the trial may have been a sicker population with a more rapid progression of disease through adjuvant chemotherapy, whereas second- and third-line patients needed to live long enough with a good enough performance status to be enrolled in the study," she said.
Another theory impacting survival data in first-line patients is that their chemotherapy was a taxane-based regimen, which could enrich for cancers benefitting from DNA-damaging agents, O'Shaughnessy added.
The Phase III trial did confirm the safety of iniparib compared to chemotherapy, as there were similar toxicities seen in patients randomized to either arm. Neutropenia, anemia, and thrombocytopenia were the most common grade 3/4 toxicities in the study, and only one patient died from complications due to treatment with the iniparib/chemotherapy regimen.
Have topics you'd like to see covered in Pharmacogenomics Reporter? Contact the editor at tray [at] genomeweb [.] com.