Skip to main content
Premium Trial:

Request an Annual Quote

PARP Inhibitor, Chemo Combo Shows Potential as Neoadjuvant Option for BRCA-Mutant Breast Cancer


This article has been updated to clarify that the BRCA1/2-positive cohort included both ER-positive and triple-negative breast cancer patients.

SAN DIEGO – Readouts from the Phase II/III PARTNER trial demonstrated that a PARP inhibitor and chemotherapy may be efficacious and more tolerable as neoadjuvant treatment if given two days apart in breast cancer patients with BRCA1/2 mutations.

The same neoadjuvant treatment strategy, however, was not as efficacious in TNBC patients who had BRCA1/2 wildtype tumors, though these tumors often behave like they have homologous recombination repair deficiencies.  

The PARTNER trial, which has been ongoing in the UK since 2016, evaluated AstraZeneca's PARP inhibitor Lynparza (olaparib) plus chemotherapy against chemo alone as a neoadjuvant treatment across different breast cancer genomic subgroups. At the American Association for Cancer Research annual meeting on Monday, researchers presented data from two cohorts of this trial: patients with germline BRCA1/2 mutations and TNBC patients with BRCA1/2 wildtype tumors.

The two cohorts had starkly different outcomes on Lynparza-chemo. All patients in the BRCA1/2-mutant group were still alive after three years on the Lynparza-containing regimen, whereas the combination failed to beat chemo in the BRCA1/2-wildtype group.

In this study, researchers also explored a unique dosing schedule for the Lynparza-chemo combination by giving Lynparza 48 hours after chemo. They hypothesized that giving Lynparza and chemo with a gap would be less toxic than giving them together and tested whether the order in which they were given made a difference in terms of safety.

In the UK, Lynparza is available as an adjuvant treatment for BRCA1/2 mutation-positive, HER2-negative high-risk early breast cancer after chemotherapy. The UK's National Institute for Health and Care Excellence (NICE) is also considering whether the National Health Service should make the PARP inhibitor available to BRCA1/2-positive HER2-negative metastatic breast cancer patients as a second-line treatment. In the US, the drug is approved in the same adjuvant setting and in the second-line settings.

BRCA1/2-mutant cohort

Among 86 breast cancer patients with BRCA1/2 mutations in the PARTNER trial, pathologic complete response rates were similar for those who got Lynparza 48 hours after chemo and the control arm, 64.1 percent and 69.8 percent, respectively. The BRCA1/2 cohort included patients with either triple-negative or ER-positive HER2-negative breast cancer. However, patients on Lynparza-chemo had excellent survival, Jean Abraham, a professor in the department of oncology at Cambridge University, said while presenting the PARTNER trial results. "This was a significant landmark for these patients," Abraham said.

At three years, all patients who received Lynparza-chemo were alive compared to 82.8 percent of patients in the chemo alone arm. Only one patient experienced cancer recurrence in the Lynparza-chemo arm, translating to a three-year event-free survival rate of 96 percent. Comparatively, in the control arm, nine patients saw their cancers return, translating to an 80 percent three-year event-free survival rate.

When researchers looked at the extent to which pathologic complete response correlated with overall survival, they found that for patients in the control arm, if you didn't achieve pathologic complete response, you had worse survival. "But in the research arm, actually, it didn't seem to matter whether you had a pathological complete response. You still survived," Abraham noted.

Abraham theorized that the lack of an association in the BRCA1/2-mutated group between pathologic complete response and overall survival in the Lynparza arm may be due to the fact that cancer cells remaining after treatment are still being counted as residual disease, even though the cells may not have metastatic potential.

Researchers may need to reconsider what a non-pathological complete response means in the context of neoadjuvant treatment, she suggested. "We're not the first trial to show that there is a disconnect between non-pathological complete response and the relationship with survival," she said. "We need to understand what that means. Is it treatment context dependent? And what does it mean for [designing] adjuvant studies?"

The PARTNER trial also explored another dosing schedule, in which they gave Lynparza 48 hours before chemo, but patients on this regimen had worse outcomes than patients who received chemo first and then Lynparza and those who received just chemo.

This 48-hour dosing schedule was inspired by preclinical research done by an AstraZeneca researcher, Abraham noted. That research suggested that there was reduced bone marrow toxicity and increased antitumor activity if Lynparza was given separately from chemo.

The trial investigators decided to integrate this preclinical knowledge into the design of the PARTNER study. One arm would explore the effect of giving Lynparza 48 hours after chemo and another would test what happened when Lynparza treatment came 48 hours before chemo. The thought behind testing both these treatment sequences was to see whether the nausea and vomiting side effects from chemo would prevent some patients from taking Lynparza, Abraham explained. The investigators later dropped the Lynparza-first arm after an initial safety and efficacy analysis showed that Lynparza after chemo had better outcomes. Ultimately, most patients received at least 80 percent of the planned Lynparza dose when given after chemo.

"One of the major problems has been the compounding bone marrow toxicity that we find when we bring together PARP inhibitors and carboplatin-containing regimens," Abraham said, adding that safety data from this study showed the 48-hour gap between chemo and Lynparza was a "relatively well-tolerated regimen."

She concluded that giving Lynparza and chemo with a 48-hour gap in between was a "potentially curative regimen" for this germline BRCA1/2-mutant early-stage breast cancer group and added that the treatment should be confirmed in a larger study.

"For novel agents, we need to think about the scheduling because clearly, scheduling was really important in this trial," Abraham said. "So, this may be true of other PARP inhibitors and potentially other DNA damage repair agents and beyond."

In a discussion of the data from PARTNER, Hope Rugo, director of the breast oncology clinical trials program at the University of California, San Francisco, noted that this study was small and closed with less than half the number of patients researchers had planned to enroll. This, in turn, may have affected the researchers' ability to assess other characteristics that may affect treatment outcomes or cause toxicities, she noted.

This "does highlight the difficulty we have in enrolling patients in trials where we're focusing on BRCA1/2 mutations because you do need the rapid testing, which we can generally do in the US but not in many places in the world," Rugo said.

Abraham acknowledged the challenges of enrolling BRCA1/2-mutant patients in the PARTNER trial due to the lack of rapid upfront testing for BRCA1/2 in the UK to identify these patients and challenges related to the COVID-19 pandemic. She added that the investigators stopped recruitment in the study after the UK made Lynparza available as an adjuvant treatment for breast cancer, which she explained would have affected survival outcomes in the PARTNER trial.

BRCA1/2 wildtype patients

While Lynparza is often used as a treatment for patients with various tumors harboring a germline BRCA1/2 mutation, the PARTNER trial investigators also explored the drug in BRCA1/2 wildtype TNBC patients because TNBC can behave like BRCA1/2-mutated tumors, said Karen Pinilla, a clinical research fellow at the Cambridge Breast Cancer Research Unit, who presented the results at AACR from the wildtype cohort in the PARTNER trial.

"Triple-negative tumors that are wildtype for these mutations frequently have similarities to BRCA1/2-mutated tumors, with a good proportion of these [tumors] showing homologous recombination deficiency resulting in a BRCA-like phenotype," Pinilla said. "The question is, could PARP inhibitors work in this subgroup?"

This cohort, involving 550 patients with BRCA1/2 wildtype basal-like early-stage TNBC, were randomized to the same experimental 48-hour gap schedule for Lynparza and chemo or just chemo. There was no significant difference in pathological complete response rates between the Lynparza-chemo arm and the chemo alone arm, 51.1 percent versus 52.4 percent, respectively.

However, unlike what was seen among those with BRCA1/2 mutations, Lynparza-chemo didn't significantly improve survival over chemo for wildtype patients. The three-year event-free survival rate was 80 percent in the Lynparza-chemo arm versus 79 percent in the control arm. Similarly, the three-year overall survival rate was 90 percent and 87 percent, respectively, which was not a statistically significant difference.

Also counter to the BRCA1/2-mutated group, the wildtype group did show a "strong association" between survival outcomes and complete response. Among those who achieved a complete response on Lynparza-chemo, the three-year survival rate was 96 percent, compared to 83 percent for those who did not achieve a complete response.

Between the BRCA1/2-mutant and BRCA1/2-wildtype groups, the rate of complete response was significantly different, Pinilla said, adding that "this intrinsic differential response [between] the cohorts really highlights the huge heterogeneity that is seen in this group of patients and the importance of patient selection when we are designing clinical trials."

The PARTNER trial also includes a translational research component that will allow Abraham's group to continue exploring this data. The ongoing research involve efforts to better understand the heterogeneity of TNBC, to identify potential biomarkers of response to PARP inhibitors, and to discover new treatment strategies for this group.

"Despite the negative results in [the wildtype] cohort, the gap schedule regimen [for Lynparza and chemo] provides a biologically relevant concept that should be considered when testing future PARP inhibitor combinations," Pinilla said.

The overall data from PARTNER suggest a potential opportunity for PARP inhibitor-chemo combinations in the neoadjuvant BRCA1/2-mutant TNBC setting, Rugo said. While the standard-of-care neoadjuvant therapy today for TNBC patients is typically immunotherapy, there can be significant, lifelong toxicities from that treatment, she said.

Rugo echoed the need to better understand the role of PARP inhibitors in this setting, including improved strategies for decreasing the toxicity from these drugs. "Next-generation PARP1 selective inhibitors have been met with great excitement for this," Rugo said. "PARP2 is essential for hematopoiesis, so we could potentially reduce the hematologic toxicity and enable combinations."

Also at AACR this week, researchers presented data from AstraZeneca's investigational PARP1 inhibitor saruparib in which the drug demonstrated responses in nearly half of previously treated hormone receptor-positive and triple-negative breast cancer patients harboring mutations in homologous recombination repair genes in a Phase I/II trial. There was also data from trials of at least eight different PARP1 selective inhibitors presented at the meeting, showing the flurry of activity in the PARP1 space.