Neoadjuvant HER3-DXd Maintains Response Rates with Fewer TRAEs in Breast Cancer

For patients with HR–positive/HER2-negative breast cancer, neoadjuvant patritumab deruxtecan with or without letrozole showed antitumor activity.

Comparable pathologic complete response (pCR) and objective response rate (ORR) with more favorable treatment-related adverse effect (TRAE) data was attained by neoadjuvant patritumab deruxtecan (HER3-DXd) with or without letrozole (Femara) compared with multiagent chemotherapy, based on results from the three-arm, randomized, open-label phase 2 SOLTI VALENTINE trial (NCT05569811).¹

Data, which were presented during the 2024 San Antonio Breast Cancer Symposium (SABCS), showed that the antibody-drug conjugate (ADC; n = 50) elicited a pCR rate of 4.0% (95% CI, 0.5%-13.7%) and an ORR of 70.0% (95% CI, 55.4%-82.1%). When paired with letrozole (n = 48), the pCR rate was 2.1% (95% CI, 0.1%-11.1%) and the ORR was 81.3% (95% CI, 67.4%-91.1%); 1 patient experienced progressive disease (PD). With standard multiagent chemotherapy (n = 24), the pCR rate was 4.2% (95% CI, 0.1%-21.1%), and the ORR was 70.8% (95% CI, 48.9%-87.4%); 1 patient had PD. In the total population of 122 patients, the pCR rate was 3.3% (95% CI, 0.9%-8.2%), and the ORR was 74.6% (95% CI, 65.9%-82.0%); 2 patients had PD.

Moreover, patritumab deruxtecan demonstrated biological evidence of antitumor activity, with a drop in Ki67, a switch to less proliferative PAM50 subtypes, and a decrease in risk of recurrence (ROR). The ADC also led to an increase in CelTIL score that correlated with treatment response.

“SOLTI VALENTINE supports the effectiveness of [patritumab deruxtecan] in early breast cancer and its potential incorporation in the treatment of high-risk HR-positive/HER2-negative breast cancer,” Mafalda Oliveira, MD, PhD, of Vall d’Hebron University Hospital, Vall d’Hebron Institute of Oncology, and of the SOLTI Cancer Research Group, both in Barcelona, Spain, said in a presentation of the data.

It is known that multiagent chemotherapy regimens boost survival outcomes for this population, but a need remains for not only effective, but tolerable, therapeutic approaches. The first-in-class, HER3-directed ADC patritumab deruxtecan has demonstrated activity in several subtypes of breast cancer. Previously, a single dose of the agent resulted in increased CelTIL score and boosted clinical response in this population enrolled to the window-of-opportunity early phase 1 SOLTI TOT-HER3 study (NCT04610528).^2,3

Based on these observations, investigators hypothesized that patritumab deruxtecan monotherapy or in combination with letrozole would be a safe and effective neoadjuvant option for treatment-naive patients with high-risk, HR-positive/HER2-negative early breast cancer.¹

To test this, they launched the parallel, randomized, noncomparative, open-label, phase 2 SOLTI VALENTINE trial, which sought to enroll 120 pre- and postmenopausal patients with primary operable breast cancer of 1 cm or larger by MRI. To be eligible, patients needed to have HR-positive/HER2-negative disease, as well as a Ki67 expression of 20% or higher and/or high genomic risk. They could not have previously received treatment for their current breast cancer. They also needed to have pretreatment formalin-fixed paraffin-embedded core-needle biopsy.

Patients were randomly assigned 2:2:1 to receive patritumab deruxtecan at 5.6 mg/kg every 21 days for 6 cycles; patritumab deruxtecan at the same dose plus letrozole at 2.6 mg once daily with or without ovarian function suppression; or standard multiagent chemotherapy in the form of epirubicin at 90 mg/m² plus cyclophosphamide at 600 mg/m² or doxorubicin at 60 mg/m² plus cyclophosphamide at 600 mg/m² for 4 cycles followed by weekly paclitaxel at 80 mg/m² for 12 weeks. Patients were stratified by axillary nodal status. Patients then went on to surgery and will be followed for 5 years.

The primary end point of the study was pCR rate (ypT0/is ypN0) at surgery, and secondary end points included ORR, CelTIL score change, Ki67 drop, PAM50, ROR, and safety.

“The sample size was not selected to formally compare treatment arms in terms of pCR rate, but to achieve a certain level of precision estimating pCR in each arm,” Oliveira noted. Investigators assumed a 20% pCR rate in both ADC arms and that 48 patients per arm could provide a precision estimate of ±11.3% leveraging asymptotic 2-sided 95% confidence intervals. They included the chemotherapy arm to serve as an internal response control and to ensure the correct population was recruited to the study.

The interim analysis would happen when there were 30 evaluable patients and they would evaluate pCR, ORR, and safety. The final analysis will happen when 120 patients are evaluable, and they will report data on the primary end point. In the long-term follow-up, they will conduct an invasive disease–free survival analysis.

Patients were recruited between November 2022 and September 2023. A total of 154 patients were assessed for eligibility, and 32 were excluded due to withdrawn consent (n = 8), metastatic disease (n = 9), no pretreatment biopsy (n = 2), and other (n = 13). A total of 122 patients were included in the analysis; 50 received the ADC alone, 48 received the doublet, and 24 received chemotherapy; 49, 48, and 23 patients, respectively, went on to surgery. The data cutoff date for the analysis shared during the meeting was April 22, 2024.

In the overall population, the median age was 51 years (range, 29-82), and 99.2% were female. About half (52.9%) were premenopausal, and most had lymph node positivity (76.2%). More than half of patients had stage II disease (64.0%) and grade 1 or 2 disease (78.4%). Regarding cT stage, 42.0% of those in the ADC-alone arm, 39.6% of those in the ADC/letrozole arm, and 38.5% of those in the chemotherapy arm had T3-T4 disease. Overall, the median local Ki67 expression was 35 (range, 18-90). Intrinsic subtypes (PAM50) included basal-like (0.8%), HER2-enriched (3.3%), luminal A (37.2%), luminal B (55.4%), or nominal-like (3.3%); this information was not available for 1 patient.

Additional data showed that in the ADC arms, there was a significant change in CelTIL score from baseline to day 1 of cycle 2, and from baseline to surgery. Oliveira added that a link between CelTIL change from baseline to day 1 of cycle 2 and radiological response was observed. Lastly, there was a shift from high or medium risk of recurrence scores at baseline to low scores at the time of surgery; this was also true for the chemotherapy arm.

Regarding safety, any-grade TRAEs occurred in 96.0% of those in the ADC-alone arm (n = 50), 97.9% of those in the ADC/letrozole arm (n = 48), and 95.8% of those in the chemotherapy arm (n = 24); they were grade 3 or higher in 14.0%, 14.6%, and 45.8% of patients, respectively. In the ADC-alone arm, dose reduction and interruption were needed for 6.0% and 12.0% of patients, respectively. No patients in this arm experienced TRAEs that led to treatment discontinuation. In the ADC/letrozole arm, dose reduction, interruption, and treatment discontinuation was needed for 8.3%, 4.2%, and 4.2% of patients, respectively; in the chemotherapy arm, these respective rates were 37.5%, 50.0%, and 8.3%.

No cases of interstitial lung disease were reported, and no treatment-related deaths occurred. However, one non-related death occurred in the ADC arm due to a cerebrovascular accident.

The most common non-hematologic TRAEs reported in 20% or more of patients in the ADC-alone, ADC/letrozole, and chemotherapy arms included fatigue (50.0%; 68.8%; 79.2%), nausea (56.0%; 72.9%; 58.3%), alopecia (52.0%; 68.8%; 45.8%), diarrhea (42.0%; 54.2%; 8.3%), stomatitis (14.0%; 18.8%; 41.7%), increased transaminase levels (14.0%; 37.5%; 20.8%), neurotoxicity (0%; 0%; 33.3%), abdominal pain (16.0%; 29.2%; 8.3%), constipation (20.0%; 29.2%; 8.3%), and vomiting (22.0%; 22.9%; 8.3%).

The most common hematologic TRAEs in the respective arms included neutropenia (26.0%; 18.8%; 41.7%), anemia (12.0%; 14.6%; 33.3%), lymphopenia (2.0%; 4.2%; 8.3%), thrombocytopenia (2.0%; 4.2%, 12.5%), leukopenia (4.0%; 0%; 8.3%), and febrile neutropenia (0%; 0%; 4.2%).

Grade 3 or higher TRAEs were less frequent in the ADC arms, Oliveira concluded.

Disclosures: Dr Oliveira is a consultant for AstraZeneca, Cureos Science, Daiichi Sankyo/AstraZeneca, Gilead, iTEOS, Lilly, MSD, ProteinQure, Relay Therapeutics, Roche, and Seagen. Grant or research support to the institution was received from AstraZeneca, Ayala Pharmaceuticals, Boehringer-Ingelheim, Genentech, Gilead, GSK, Immutep, Roche, Seagen, and Zenith Epigenetics. Honoraria was received from AstraZeneca, Eisai, Gilead, Libbs, Lilly, MSD, Novartis, Pfizer, Roche, and Seagen. Travel grants were received from AstraZeneca, Eisai, Gilead, Pierre-Fabre, and Roche. She was also president elected for SOLTI Breast Cancer Group.

References

1. Oliveira M, Pascual T, Parraga KA, et al. Neoadjuvant HER3-DXd alone or in combination with letrozole for high-risk HR+/HER2- early EBC: Primary results of the randomized phase II SOLTI VALENTINE trial. Presented at: 2024 San Antonio Breast Cancer Symposium; December 10-14, 2024; San Antonio, TX. Abstract LB1-06.
2. Oliveira M, Falato C, Cejalvo JM, et al. Patritumab deruxtecan in untreated hormone receptor-positive/HER2-negative early breast cancer: final results from part A of the window-of-opportunity SOLTI TOT-HER3 pre-operative study. Ann Oncol. 2023;34(8):670-680. doi:10.1016/j.annonc.2023.05.004
3. Brasó-Maristany F, Ferrero-Cafiero JM, Falato C, et al. Patritumab deruxtecan in HER2-negative breast cancer: part B results of the window-of-opportunity SOLTI-1805 TOT-HER3 trial and biological determinants of early response. Nat Commun. 2024;15(1):5826. doi:10.1038/s41467-024-50056-y