Germline and somatic oncogenic alterations are frequently seen in breast cancer, yet it’s not well understood how their interaction influences tumor evolution and therapy resistance. Researchers now have new insights through a comprehensive integrated clinicogenomic analysis and preclinical investigation led by Memorial Sloan Kettering Cancer Center (MSK) and published in Nature on March 4, 2026.
The MSK-led multi-institutional study combined large-scale clinical-genomic analyses with mechanistic laboratory modeling to define how inherited and tumor-acquired alterations interact to shape therapeutic resistance. Specifically, the researchers found that patients with germline BRCA2 gene mutations or homologous recombination deficiency (HRD) from other causes experienced worse outcomes on the current frontline standard-of-care in the metastatic HR+/HER2– setting, a combination of a cyclin-dependent kinase 4/6 (CDK4/6) inhibitor, plus endocrine therapy (ET). The findings were validated using large real-world multi-institutional datasets representing the general metastatic breast cancer patient population treated in both community and academic settings, as well as genomic data from a pivotal Phase 3 clinical trial.
The strength and translational implications of these findings directly informed the design of the global Phase 3 clinical trial, with the paper’s corresponding author, Pedram Razavi, MD, PhD, a physician-scientist and Director of the MSK Breast Translational Program and MSK Translational Oncology Partnership Program serving as Global Principal Investigator.
“For the first time, we have been able to predict how a subset of breast cancers evolve under therapeutic pressure and develop a specific mechanism of resistance,” said Dr. Razavi. “There are multiple pathways for breast cancer to become resistant to CDK4/6 inhibitors, but we identified a group of patients whose tumors are predisposed to develop resistance through acquiring RB1 gene alterations under the selective pressure of CDK4/6 inhibition, and importantly, that risk is predictable based on baseline genomic characteristics.”
Study Design and Findings
The study drew on data from the MSK Breast Translational Program and was overseen by Dr. Razavi with analyses led by first-author Anton Safonov, MD. The study was conducted with a multidisciplinary team of MSK investigators and international collaborators, and integrated detailed clinical annotation with sequencing data from 6,927 tumors in 5,881 breast cancer patients who underwent paired tumor and germline sequencing using the FDA-authorized MSK-IMPACT assay. The team first confirmed that the cohort was representative of the broader population of patients with breast cancer and then performed a comprehensive analysis of interactions between inherited and tumor-specific alterations to understand how these factors jointly influence clinical outcomes and therapeutic resistance. The researchers showed that patients with pathogenic germline BRCA2 variants were significantly more likely to develop somatic RB1 loss-of-function mutations and experienced poorer outcomes on CDK4/6 inhibitor-based frontline therapy. They validated their findings using an independent, nationwide clinicogenomic dataset from patients with HR+/HER2– metastatic breast cancer (MBC), treated with first-line CDK4/6 inhibitor combinations, in community oncology and academic medical center settings. Additional analysis revealed that germline BRCA2 status did not universally determine poor outcomes, but rather offered context-dependent relevance to specific therapies, particularly CDK4/6 inhibitor-based therapy.
Researchers then focused on HR+/HER2– MBC patients in the MSK cohort who received a PARP inhibitor after progression on CDK4/6 inhibitor-based therapy. Despite being offered in later lines of treatment, PARP inhibitors generally resulted in significantly improved outcomes compared to the CDK4/6 inhibitor regimen. The clinical benefit was particularly more pronounced in patients who failed to respond to prior CDK4/6 inhibitor combinations.
A key mechanistic clue emerged from the genomic architecture: RB1 and BRCA2 are both located on chromosome 13q, and tumors with germline BRCA2 mutations frequently undergo loss of heterozygosity (LOH) at BRCA2. This event often results in concurrent hemizygosity (only one functional copy) of RB1, effectively priming the tumor for acquisition of a second RB1 hit. Based on this hypothesis, the team compared the clinical relevance of pre-treatment RB1 hemizygosity and other allelic configurations, including loss prior to or post whole-genome doubling, which results in more than one remaining wild-type RB1 allele. They found that only pre-treatment RB1 hemizygosity uniquely predisposed the patient to acquiring RB1 loss-of-function (LoF) when exposed to CDK4/6 inhibitors. They also found that acquired-RB1 LoF occurred almost exclusively in this setting, while rare in other allelic settings.
They concluded that RB1 hemizygosity lowers the evolutionary barrier to resistance, requiring only one additional hit for complete inactivation. In contrast to Knudson’s classical two-hit model— in which copy number loss serves as the necessary second hit for oncogenesis in patients born with an inherited defective RB1 allelethese data support an inverted paradigm, where RB1 hemizygosity increases the likelihood of developing resistance, requiring only one additional hit (generally with a LoF mutation) for a complete inactivation under CDK4/6 inhibitor selective pressure. The team further demonstrated that HRD independently contributes to this process by increasing the probability of acquiring RB1 LoF mutations.
Dr. Razavi and his team collaborated with colleagues in Europe to further validate the findings, including Dr. Nicolas Turner and Dr. Ben O’Leary at The Royal Marsden Hospital and the Institute of Cancer Research in London. The group confirmed the results using genomic data from the pivotal Phase 3 PALOMA-3 clinical trial.
Dr. Safonov first presented these findings at the San Antonio Breast Cancer Symposium in 2023 and later updated the results at the ASCO Clinical Science Symposium in 2024. “Small case series had suggested that BRCA mutations might be linked to resistance to CDK4/6 inhibitors, but no study had demonstrated this as definitively as we did or provided a clear biological explanation for it,” he said. Currently, patients with HR+/HER2– metastatic breast cancer receive CDK4/6 inhibitor-based therapy as first-line therapy, with PARP inhibitors used later. “These data provide a strong rationale to reconsider treatment sequencing and potentially prioritize PARP inhibitors in patients with HRD-positive disease,” Dr. Safonov added. “This strategy might even warrant investigation in high-risk early-stage breast cancer.”
Collaboration With MSK Laboratories
A major highlight of the study was the extensive preclinical work led by the paper’s co-corresponding author, physician-scientist Sarat Chandarlapaty, MD, PhD, and co-first author Minna Lee, MD, a surgeon-scientist in Chandarlapaty’s laboratory. Using patient-derived xenograft models from BRCA2-mutant breast cancers, the team demonstrated that these tumors had inferior responses to CDK4/6 inhibitor–based combination therapy, with many developing loss of Rb protein expression following treatment exposure. These preclinical findings closely mirrored and reinforced the clinicogenomic analyses observed in patients.
The MSK team also collaborated with researchers at Institut Curie and AstraZeneca, utilizing independent patient-derived models that demonstrated similar findings. These confirmatory studies further supported the observation that treatment with PARP inhibitors resulted in superior outcomes compared to CDK4/6 inhibitor-based therapy in HRD-positive tumors.
“This study underscores how critical it is to integrate clinical observations with rigorous laboratory modeling,” Dr. Chandarlapaty said. “The ability to test hypotheses generated from patient data in well-annotated patient-derived models and engineered cell lines allows us to move beyond correlation and establish biological causality. Only through this level of mechanistic rigor can we gain the confidence needed to translate these insights back into the clinic and design trials that meaningfully change patient care.”
A Patient-Centered Path to Discovery
The team expressed profound gratitude to thousands of patients who participated in MSK’s translational research programs and contributed valuable clinical and genomic data. Their partnership and trust made these large-scale analyses possible, allowing MSK researchers to connect real patient experiences with meaningful biological insights and, ultimately, translate those discoveries back to clinic through more precise, biologically-informed treatment strategies.
Dr. Razavi noted that one of the most meaningful aspects of the study was the use of patient-derived models from a patient who admirably participated in MSK’s Last Wish Program, providing critical tumor samples after her passing. These samples were instrumental in validating the researchers’ hypothesis and demonstrated that restoration of homologous recombination function through BRCA2 reversion mutations could resensitize tumors to CDK4/6 inhibitors. This provided additional biological rationale for prioritizing PARP inhibitors over CDK4/6 inhibitors in patients with BRCA2-associated disease.
“The inspiration for this work truly came from the clinic,” Dr. Razavi said. “Indeed, my patient who participated in our Last Wish Program was among the very first cases that led us to formulate this two-hit hypothesis — linking homologous recombination deficiency to RB1 loss under CDK4/6 inhibition. Their contribution helped shape the scientific direction of this study.”
Clinical Implications
The ultimate goal of translational research is to convert biological discoveries into practice-changing clinical trials. In this case, the strength and consistency of the data directly informed the launch of a global Phase 3 registration trial (EvoPAR-Breast01 trial), which is now actively enrolling patients worldwide.
“There are very few examples where translational data are compelling enough to move directly into a Phase 3 study without earlier proof-of-concept trials, and to do so even before the data are formally published,” Dr. Razavi said. The research suggests not only that patients with HRD-positive tumors may benefit more from PARP inhibitors than from CDK4/6 inhibitor–based therapy, but also that certain resistance mechanisms to PARP inhibitors may restore sensitivity to CDK4/6 inhibitors. This provides a biological rationale for sequencing therapies differently to improve long-term outcomes. The EvoPAR-Breast01 trial is testing this strategy by randomizing patients with HR+/HER2– advanced breast cancer harboring BRCA1, BRCA2, or PALB2 mutations to one of three treatment arms: saruparib (a selective PARP1 inhibitor) plus camizestrant (an oral selective estrogen receptor degrader), standard CDK4/6 inhibitor plus endocrine therapy, or CDK4/6 inhibitor plus camizestrant.
Dr. Razavi emphasized the importance of strong academic–industry collaboration in advancing translational research. He noted that AstraZeneca recognized the robustness of the data and importantly, shared the vision to rapidly advance this strategy into a global randomized Phase 3 trial, which is the essential step in bringing these findings to patients.
Future Direction
Dr. Razavi said that MSK is developing a program to forecast not only outcomes, but the mechanism of resistance. “We think if we look deep and understand the biology of the tumors better, we can potentially predict their behavior on treatment,” he said. This could possibly help clinicians anticipate and address the issue before resistance occurs. “That’s exactly what we are doing with the trial—we are trying to intercept and change the treatment before resistance. If we can expand on this idea of forecasting the mechanism of resistance, we might be able to improve our outcomes and even increase our cure rates.”
Learn more about MSK clinical trials for patients with breast cancer.
Dr. Razavi has received institutional grant/funding from Grail, Novartis, AstraZeneca, Invitae, Biothernostics, Tempus, Neogenomics, Guardant Health, Personalis, Myriad Genetics, Foresight Diagnostics, Natera, Biodesix, SAGA Diagnostics, SOPHiA Genetics, Haystack, Roche and consultation/ad-board/honoraria from Novartis, AstraZeneca, Lilly/Loxo, Stemline Therapeutics, Prelude Therapeutics, Neogenomics, Regor Pharmaceuticals, SAGA Diagnostics, SOPHiA Genetics, Tempus, Myriad Genetics, Foresight Diagnostics, Natera, Pathos AI, BioNTech.
Dr. Chandarlapaty has received institutional grant/funding from Daiichi-Sankyo and AstraZeneca, share options in Totus Medicines, and consultation/ad board/honoraria from Daiichi-Sankyo, AstraZeneca, Lilly, Casdin Capital, Merck, and Pathos AI.
Dr. Anton Safonov has received travel support from Novartis.