Phase I Study and Cell-Free DNA Analysis of T-DM1 and Metronomic Temozolomide for Secondary Prevention of HER2-Positive Breast Cancer Brain Metastases

Abstract

Purpose: Preclinical data showed that prophylactic, low-dose temozolomide (TMZ) significantly prevented breast cancer brain metastasis. We present results of a phase I trial combining T-DM1 with TMZ for the prevention of additional brain metastases after previous occurrence and local treatment in patients with HER2+ breast cancer.

Patients and methods: Eligible patients had HER2+ breast cancer with brain metastases and were within 12 weeks of whole brain radiation therapy (WBRT), stereotactic radiosurgery, and/or surgery. Standard doses of T-DM1 were administered intravenously every 21 days (3.6 mg/kg) and TMZ was given orally daily in a 3+3 phase I dose escalation design at 30, 40, or 50 mg/m2, continuously. DLT period was one 21-day cycle. Primary endpoint was safety and recommended phase II dose. Symptom questionnaires, brain MRI, and systemic CT scans were performed every 6 weeks. Cell-free DNA sequencing was performed on patients' plasma and CSF.

Results: Twelve women enrolled, nine (75%) with prior SRS therapy and three (25%) with prior WBRT. Grade 3 or 4 AEs included thrombocytopenia (1/12), neutropenia (1/12), lymphopenia (6/12), and decreased CD4 (6/12), requiring pentamidine for Pneumocystis jirovecii pneumonia prophylaxis. No DLT was observed. Four patients on the highest TMZ dose underwent dose reductions. At trial entry, 6 of 12 patients had tumor mutations in CSF, indicating ongoing metastatic colonization despite a clear MRI. Median follow-up on study was 9.6 m (2.8-33.9); only 2 patients developed new parenchymal brain metastases. Tumor mutations varied with patient outcome.

Conclusions: Metronomic TMZ in combination with standard dose T-DM1 shows low-grade toxicity and potential activity in secondary prevention of HER2+ brain metastases.

Conflict of interest statement

SJ, WZ, SMS, NH, TSA, EB, DS, RS, CJP, BM; OA; WDF; TRM; EV; MRG; IK, SL, DN, XW declare no potential conflicts of interest.

PSS receives research funding from Daichi-Sanyo/AstraZeneca.

PKB has consulted for Tesaro, Angiochem, Genentech-Roche, ElevateBio, Eli Lilly, SK Life Sciences, Advise Connect Inspire (ACI), Pfizer, Voyager Therapeutics, Sintetica, and Dantari, received institutional research funding (to MGH) from Merck, Mirati, Eli Lilly, BMS, and Pfizer, and has received honoraria from Merck, Pfizer, and Genentech-Roche

CKA has received research funding PUMA, Lilly, Merck, Seattle Genetics, Nektar, Tesaro, G1-Therapeutics, ZION, Novartis, Pfizer; compensated consultant role for Genentech (1/2019-), Eisai (1/2019-), IPSEN (2/2019 - ), Seattle Genetics (11/15/2019 - 11/15/2020); Astra Zeneca (3/2020 -), Novartis (5/2020 - 5/2022), Immunomedics (10/1/2020 - 9/22/2021), Elucida (9/2020), Athenex (2/2021 - 2/2023); and royalties from UpToDate, Jones and Bartlet.

RC has received an unrestricted educational grant from Pfizer, and research funding to institution for clinical trials from MSD Ireland, Pfizer, Daichii Sankyo, and Astra Zeneca.

CT have received consultant and speaker fees from Seagen.

KLS has received research funding (to institution) from Pfizer and her husband has stock in Abbott Labs and Abbvie.

ASZ has received speaker fees from Medscape.

©2023 American Association for Cancer Research.

Figures

Figure 1.. Time on Study and Outcomes.

Patients received a standard regimen on 3.6 mg/m2 of T-DM1 iv. q21days and one of three indicated doses of oral TMZ (30, 40 or 50 mg/m2, dose levels 1-3, respectively), daily for each 21-day cycle. Patient number is indicated to the left. Initial local treatment and status as of 10/21 data cutoff is shown. Only two patients developed new parenchymal brain metastases over a median of 9.6 months. SRS, stereotactic radiation surgery prior to entering the study; WBRT, whole brain radition therapy prior to entering the study; PD, progressive systemic disease, LMD, leptomeningeal disease. # >20 lesions at baseline. * Radionecrosis confirmed after patient off study for suspected PD that was not confirmed.

Figure 2.. Whole-exome sequencing of plasma and CSF cfDNA samples from HER2+ breast cancer patients.

(A) Schematic of phase I study with regard to samples collected. The sample collection time points of four plasma samples and two CSF samples are labelled as “+”. (B-D) Identification of genomic alternations in the plasma and CSF cfDNA samples by Patient Number (Fig. 1). For each sample genomic alterations were compared to known variants for cancer-related genes in the MSK-IMPACT panel (B), and main pathogenic/likely pathogenic for all diseases in the ClinVar database (C). The potential driver mutations were identified using the OncodriveCLUST method (D). P, plasma; CSF, cerebrospinal fluid. The bar graph above each panel totals the number of mutations detected in each cfDNA sample. On the left, the percentage of cfDNA reads with mutations in the genes is listed. On the right, the distribution of mutation type per gene is indicated. On the bottom of each panel, blue lines identify two patients who developed new parenchymal brain metastases during this trial, while red lines identify four long-term survivors (>12 months without a new parenchymal brain metastasis).

Figure 3.. Mutation patterns for patients with new brain metastases versus long-term responders.

Potential relationship between mutations in plasma and CSF cfDNA throughout the study, and the patients’ clinical outcome. Two patients who developed new brain metastases during this trial are listed as “New Brain Metastasis”, while four patients who remained free of a new parenchymal brain metastasis for at least 12 months were identified as “Long-term survivors” (see Figure 1). The type of mutations is indicated by a colored circle.