Durvalumab Plus Trastuzumab/Pertuzumab Elicits Responses in HER2-Enriched Breast Cancer

Administration of a neoadjuvant, chemotherapy-free regimen comprising trastuzumab (Herceptin) plus pertuzumab (Perjeta) and durvalumab (Imfinzi) was tolerable and produced high pathologic response rates comparable to those achieved with a standard salvage chemotherapy regimen in patients with HER2-enriched early breast cancer, according to data from a prospective, open-label, phase 2 trial (NCT03820141) presented during the 2024 AACR Annual Meeting.¹

Among all evaluable patients (n = 37), 67.6% achieved a pathologic response and 48.6% achieved a pathologic complete response (pCR) with the de-escalated treatment regimen. Notably, half of non-responders (n = 6) went on to receive the standard chemotherapy regimen, 50% of whom achieved a pCR. With the addition of these 3 patients, the overall pathologic response rate increased to 75.7% and the overall pCR rate increased to 56.8%.

“Our de-escalated, chemotherapy-free regimen was well tolerated in the HER2-enriched population and this regimen was able to [produce] high pCRs,” said lead study author Jian Guan, MD, PhD, who is a fellow at Houston Methodist Neal Cancer Center in Texas, in an oral presentation of the data. “Even for the patients initially not responding to the treatment, the salvage [chemotherapy] regimen yielded a very reasonable pCR rate.”

In high-risk HER2-positive early breast cancer, conventional treatment involves the use of neoadjuvant chemotherapy along with HER2-directed therapy. However, uncertainty surrounding the benefits of chemotherapy in the HER2-enriched subset has spurred ongoing research into the feasibility of treatment de-escalation.

“[The combination of docetaxel, carboplatin, trastuzumab, and pertuzumab] has emerged as the standard for early-stage, high-risk HER2-positive breast cancer primarily because of its success in inducing a high pCR,” Guan explained. “Nevertheless, there remains an unmet need to further improve this treatment strategy and to improve response rates. Appropriate patient selection is the key to the development of a personalized treatment plan and the prevention of overtreatment.”

Preclinical studies have shown a potential synergistic effect between immune checkpoint blockade and HER2-directed therapy, and trastuzumab is known to induce immunosuppression by upregulating PD-L1 expression in macrophages. Accordingly, this study investigated whether adding durvalumab to trastuzumab and pertuzumab would increase response rates achieved with dual HER2 blockade and potentially eliminate the need for chemotherapy in patients with HER2-enriched early breast cancer.

Accordingly, this single-arm, phase 2 study enrolled patients older than 18 years of age with stage I to III estrogen receptor–negative, HER2-positive early breast cancer who had not received prior treatment.^1,2 Patients also had an ECOG performance status of 0 or 1, no contraindications for checkpoint inhibition, a normal left ventricular ejection fraction of at least 50%, and adequate organ function.

For 6 cycles, patients in the experimental arm were treated with 1120 mg of intravenous (IV) durvalumab, an 8-mg/kg IV loading dose of trastuzumab followed by 6 mg/kg, and a 840-mg IV loading dose of pertuzumab followed by 420 mg every 3 weeks.² At the end of cycle 6, responses were assessed using breast MRI, and/or biopsy if residual disease was present. Patients who showed no residual disease proceeded to surgery, and those with biopsy-proven residual disease were offered salvage standard chemotherapy prior to surgery.

The study’s primary end point was pathologic response rates, defined as a residual cancer burden (RCB) score of 0 or 1. Key secondary end points included pathologic response rates according to a PD-L1 expression of 1% or higher or below 1% and TIL levels below 5% and 5% or higher, 3-year disease-free survival, and safety. Evaluation of correlative molecular and genetic biomarkers of response and resistance served as an exploratory objective.

Of the 51 patients screened during trial enrollment, 3 withdrew consent and 3 were withdrawn for other reasons. Moreover, 6 patients were not enriched for HER2. A total of 39 patients were enrolled onto the study. Following treatment, 2 patients were not evaluable for efficacy due to a sudden death unrelated to treatment and LOF, respectively.

The median age of patients was 55 years (range, 29-85) and over 60% of patients were minorities, according to Guan, including Hispanic (38.5%), African American (10.3%), and Asian (10.3%). The majority of patients had a variant of uncertain significance or were germline mutation negative (61.5%), and 7.7% harbored pathogenic mutations. Sixty percent, 30.8%, and 10.3% of patients had a TIL expression of up to 10%, between 11% and 59%, or 60% or higher, respectively. Ki67 expression was between 6% and 29% in 15.4% of patients and 30% or higher for 76.9% of patients.

Regarding tumor staging, 25.6%, 38.5%, 23.1%, and 12.8% of patients had T1c, T2, T3, and T4 staging, respectively. Moreover, 46.2% of patients did not have cancer in nearby lymph nodes, and 43.6%, 5.1%, and 5.1% of patients had N1, N2, or N3 cancer-containing lymph nodes, respectively.

An exploratory correlation analysis revealed that patients with smaller tumors, earlier-stage disease, and high tumor-infiltrating lymphocyte (TIL) levels were more likely to achieve a pCR with the chemotherapy-free regimen. pCRs were observed in 90%, 43%, 37.5% and 0% of patients with T1 (n = 10), T2 (n = 14), T3 (n = 8), and T4 (n = 5) disease, respectively (P = .007). Similarly, 85.7% of those with stage 1 disease (n = 7) experienced a pCR compared with 52.6% of those with stage 2 disease (n = 19) and 18.2% of those with stage 3 disease (n = 11; P = .018). Lastly, the pCR rate was 73.3% in patients with TIL levels greater than 10% (n = 15) vs 31.8% in those with a TIL levels of 10% or less (n = 22; P = .02)

Safety analysis showed that the majority of adverse effects (AEs) experienced by patients were grade 1/2, and that the durvalumab regimen was generally well tolerated. All instances of cardiac arrest, mitral valve regurgitation, and pleural effusion were grade 3 events. The most common AEs experienced in over 40% of patients were low-grade rash, fatigue, and diarrhea. No new safety concerns were observed.

Similar findings were also reported from the phase 2, prospective, multicenter, open-label, neoadjuvant Keyriched-1 study (NCT03988036), which evaluated the addition of pembrolizumab (Keytruda) to neoadjuvant trastuzumab and pertuzumab in patients with the HER2-enriched subtype.^1,3 Among the 46 patients enrolled onto this study, 52% achieved a pCR.

Guan added that, “It is important to highlight that in the Keyriched-1 trial, only 7% of patients had T3 disease and 30% of patients had nodal involvement, indicating [that this study population includes patients with] predominantly earlier-stage disease. In contrast, we have a higher proportion of patients with locally advanced disease in our study cohort.”

Guan concluded by stating that subsequent investigations of molecular biomarkers for treatment response and resistance mechanisms are ongoing.

References

1. Guan J, Sun K, Jain, D et al. Chemotherapy-free neoadjuvant regimen with durvalumab, trastuzumab and pertuzumab (DTP) in HER2-enriched early breast cancer: A prospective, open-label phase II trial. Presented at: 2024 AACR Annual Meeting; April 5-10, 2024; San Diego, CA. Abstract CT029.
2. Durvalumab with trastuzumab and pertuzumab in HER2-enriched breast cancer (DTP). ClinicalTrials.gov. Updated September 18, 2023. Accessed April 15, 2024. https://classic.clinicaltrials.gov/ct2/show/NCT03820141
3. Kuemmel S, Gluz O, Reinisch M, et al. Keyriched-1- A prospective, multicenter, open label, neoadjuvant phase ii single arm study with pembrolizumab in combination with dual anti-HER2 blockade with trastuzumab and pertuzumab in early breast cancer patients with molecular HER2-enriched intrinsic subtype. Cancer Research. 2022;82(suppl 4):PD10-11. doi:10.1158/1538-7445.SABCS21-PD10-11