Abstract OT2-06-01: Phase I/II study of T-DM1 alone versus T-DM1 and metronomic temozolomide in secondary prevention of HER2-Positive breast cancer brain metastases following stereotactic radiosurgery

Background: Brain metastases occur in up to 25-40% of HER2+ breast cancer patients. Standard treatment is limited to surgery or stereotactic radiosurgery (SRS) and/or whole brain radiation therapy (WBRT), with high levels of recurrence or progression, limiting survival and quality of life in most patients. Our group has demonstrated that low doses of temozolomide (TMZ) administered in a prophylactic, metronomic fashion can significantly prevent development of brain metastases in murine models of breast cancer. Based on these findings, we propose a secondary-prevention clinical trial.

Trial Design: Phase I/II open label study. Phase I will follow a standard 3+3 design: T-DM1 3.6 mg/kg IV every 21 days plus TMZ 30, 40 or 50 mg/m2 daily. Phase II: randomization T-DM1 3.6 mg/kg versus T-DM1 3.6mg/kg plus TMZ at recommended phase 2 dose (RP2D). Patients will undergo radiology guided lumbar puncture at baseline and after 6 weeks of treatment (C3D1) for correlative studies, brain MRI, systemic restaging CTs, and questionnaires for evaluation of symptoms and quality of life (MDASI-BT and PROMIS®) every 6 weeks.

Eligibility: HER2+ breast cancer with ≤3 brain metastases, treated with SRS and/or resection ≤6 weeks before enrollment, no leptomeningeal metastases, no previous WBRT, able to complete brain MRI with contrast evaluations, willing to undergo lumbar puncture, ECOG ≤2 and adequate organ and marrow function. HBV, HCV or HIV-positive patients are ineligible.

Specific Aims: Phase I: to identify the maximum tolerated dose (MTD) of TMZ combined with T-DM1. Phase II: to determine if the combination regimen of T-DM1 and TMZ improves the recurrence-free incidence from distant new brain metastases at one year as compared to T-DM1 alone. Biomarkers, including cell free DNA sequencing from CSF, serum and tumor block, serum markers for neuroinflammation, and patient reported outcomes, will be analyzed in an exploratory fashion.

Statistical Methods: Phase I, MTD will be identified based on the dose level at which 0 or 1 patient in 6 has a DLT. Phase II, to test whether TMZ will increase RFS from 50% to 65% at 12 months. RFS Kaplan-Meier curves will be created for each of the randomized arms and compared using a one-tailed log-rank test, with a one-sided 0.10 significance level of interest to be detected. Patients will be stratified for number of brain lesions and status of systemic metastases (controlled or not).

Target Accrual: 49 evaluable patients per arm (total 98), plus 9 to 18 patients during phase I. Trial will open in Summer 2017, at NIH in Bethesda, MD.

Contact Information: Principal Investigator: Alexandra S Zimmer, MD alexandra.zimmer@nih.gov

Citation Format: Zimmer AS, Gril B, Steinberg S, Smart D, Gilbert M, Armstrong T, Xiao L, Houston N, Biassou N, Brastianos P, Carter S, Lyden DC, Lipkowitz S, Steeg P. Phase I/II study of T-DM1 alone versus T-DM1 and metronomic temozolomide in secondary prevention of HER2-Positive breast cancer brain metastases following stereotactic radiosurgery [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr OT2-06-01.

Semantic memory in Alzheimer's disease: representativeness, ontologic category, and material

Alzheimer's disease (AD) patients with semantic memory difficulty and AD patients with relatively preserved semantic memory named pictures and judged the category membership of words and pictures of natural kinds and manufactured artifacts that varied in their representativeness. Only semantically impaired patients were insensitive to representativeness in their category judgments. AD subgroup judgments did not differ for natural kinds compared to manufactured artifacts nor for words compared to pictures. AD subgroup differences could not be explained by dementia severity, memory, reading, and visuoperception. The similarity process for relating coordinate members of a taxonomic category contributes to the normal appreciation of word and picture meaning, and this process is compromised in AD patients with semantic difficulty.

Semantic memory in Alzheimer's disease: representativeness, ontologic category, and material

Alzheimer's disease (AD) patients with semantic memory difficulty and AD patients with relatively preserved semantic memory named pictures and judged the category membership of words and pictures of natural kinds and manufactured artifacts that varied in their representativeness. Only semantically impaired patients were insensitive to representativeness in their category judgments. AD subgroup judgments did not differ for natural kinds compared to manufactured artifacts nor for words compared to pictures. AD subgroup differences could not be explained by dementia severity, memory, reading, and visuoperception. The similarity process for relating coordinate members of a taxonomic category contributes to the normal appreciation of word and picture meaning, and this process is compromised in AD patients with semantic difficulty.

Dual processing of open- and closed-class words

A series of articles in the past two decades has suggested differential processing of open- and closed-class lexical items by normal adults. Difficulties in replicating a crucial study (Bradley, 1978), however, have weakened the dual route hypothesis. We matched 16 French open-class items to 16 closed-class items for phonological structure, world length, and relative word frequency. Three agrammatic aphasics revealed strikingly more phonological errors on closed-class than open-class items. Dysfluencies were greater on closed-class items and contributed to greater overall reading time for the closed-class words, consistent with a two-route model for the production of closed- and open-class lexical items in Broca's aphasics and, thus, normals.

Language comprehension profiles in Alzheimer's disease, multi-infarct dementia, and frontotemporal degeneration

We assessed language functioning in 116 age-, education-, and severity-matched patients with the clinical diagnosis of Alzheimer's disease (AD), multi-infarct dementia (MID) due to small-vessel ischemic disease, or a frontotemporal form of degeneration (FD). Assessments of comprehension revealed that patients with AD are significantly impaired in their judgments of single word and picture meaning, whereas patients with FD had sentence comprehension difficulty due to impaired processing of grammatical phrase structure. Patients with MID did not differ from control subjects in their comprehension performance. Traditional aphasiologic measures did not distinguish between AD, MID, and FD. Selective patterns of comprehension difficulty in patients with different forms of dementia emphasize that language deficits cannot be explained entirely by the compromised memory associated with a progressive neurodegenerative illness.

Phonologic processing deficits in Alzheimer's disease

We investigated phonologic production in patients with mild to moderate Alzheimer's disease (AD) on a repetition task. AD patients produced significantly more speech errors than age-matched controls. AD patients' errors, unlike those of controls, resulted in the transformation of real words into pseudowords, occurred disproportionately in word-initial positions, and were not influenced by the phonologic environment. This pattern of errors suggests a lexical phonologic retrieval deficit in AD.