Stamford Hospital

Abstract Introduction: Abemaciclib, a CDK4/6 inhibitor, reduces the risk of recurrence in patients (pts) with high-risk HR+/HER2- early breast cancer; use is associated with toxicities, particularly diarrhea, which can impact pts’ ability to maintain dose and/or remain on therapy. TRADE (NCT06001762) is a phase 2, investigator-initiated single-arm trial designed to evaluate whether an initial dose-escalation strategy improves adjuvant abemaciclib tolerability. The study met its primary endpoint, showing the dose escalation strategy reduced treatment discontinuation and improved ability to reach/maintain the target dose of 150 mg BID at 12 weeks (wks), compared to historical controls. Here, we present 6-month follow-up results from TRADE. Methods: Pts with HR+/HER2- early breast cancer eligible for abemaciclib with adjuvant endocrine therapy were enrolled. All pts initiated abemaciclib at 50 mg BID for 2 wks, escalated to 100 mg BID for 2 wks, and then to 150 mg BID onward for a planned 2-year course. Dose escalation required the absence of ongoing grade 3/4 or persistent grade 2 adverse events (AEs). Management of AEs and dose reductions beyond 12 wks was per standard of care. Data on adherence to oral therapy and ctDNA dynamics were collected. Results: Ninety pts were enrolled; median age was 58 years (range 24-78), and the majority were White (80.0%). 51.1% had stage II disease and 48.9% stage III disease. 73.3% had received chemotherapy in the neo/adjuvant setting. All pts received an adjuvant aromatase inhibitor (AI), 17.8% with ovarian function suppression. One pt had progression within 24 wks of treatment, hence was excluded from dose-related analyses. Out of 89 evaluable pts, 72 (80.9%) remained on treatment at 24 wks; 46 (63.9%) were receiving abemaciclib at 150 mg BID, 18 (25.0%) at 100 mg BID and 8 (11.1%) at 50 mg BID. Abemaciclib discontinuation occurred in 6 pts (6.7%) during the first 12 wks and in an additional 11 pts (12.4%) between wks 12 and 24, resulting in a cumulative discontinuation rate of 19.1% (n=17) over the initial 24 wks of exposure. The most common reasons for discontinuation were AEs (7, 7.9%) and withdrawal of consent/patient preference (6, 6.7%). Specific AEs contributing to discontinuation included diarrhea (2, 2.2%), pneumonitis (2, 2.2%), transaminitis (1, 1.1%), allergic reaction (1, 1.1%), and other (1, 1.1%). Of those who discontinued for patient preference, 5 of the 6 had reached target dose, of whom 3 did not have a dose reduction. A total of 12 pts (13.5%) did not reach target dose, and 29 (32.6%) required dose reductions within 24 wks, primarily due to gastrointestinal or hematologic toxicities, 22 (24.7%) with reductions from 150 mg dosing. The most frequently reported AEs overall were diarrhea (grade ≥2 32.2%, grade 3 5.6%), neutropenia (31.1%, grade ≥3 4.4%), and fatigue (25.6%, no grade ≥3). When analyzed by 4-wk intervals, rates of grade >2 diarrhea were 8.9% (wks 1-4), 19.1% (wks 5-8), 16.5% (wks 9-12), 12.0% (wks 13-16), 8.9% (wks 17-20), and 9.5% (wks 21-24), showing an early peak and gradual decline over time. Patient-reported outcomes, measured by FACT-B assessments, indicated that quality of life remained stable throughout the study period. Additional analyses, including evaluation of adherence and ctDNA, are ongoing. Conclusions: In the TRADE study, after initial dose escalation, abemaciclib was generally well tolerated, with a cumulative discontinuation rate of 19.1% at 24 wks, 7.9% for AE. The incidence of grade >2 diarrhea peaked at month 2 and declined over time. The majority of pts maintained quality of life and were able to remain on therapy with manageable and expected toxicities. These updated findings continue to support the feasibility of an initial abemaciclib dose-escalation approach to optimize drug exposure and reduce early discontinuation, while preserving therapeutic intensity. Citation Format: I. Schlam, D. Trapani, S. Kim, M. Faggen, N. Sinclair, P. Sanz-Altamira, C. Battelli, S. Berwick, S. Lo, J. Acevedo, S. Sinclair, A. Malcom, L. Varella, S. L. Sammons, S. T. Schumer, P. D. Poorvu, E. P. Wallace, E. Pasternak, N. Tayob, S. M. Tolaney, E. L. Mayer. Trade: 6-month abemaciclib tolerability after initial dose escalation in early HR+/HER2- breast cancer [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PD1-08.

Abstract Introduction: Neoadjuvant chemotherapy (CT) with pembrolizumab (P) is standard for most patients (pts) with high-risk early TNBC. Biomarkers for the addition of P to CT are lacking. The prognostic and predictive value of TILs in pts who receive neoadjuvant CT+P vs CT alone and the significance of change in TILs from pre- to post-treatment samples remains unclear. Methods: From the prospective DFCI Multicenter TNBC Registry, pts with stage I-III TNBC who received neoadjuvant CT or CT+P were identified. Stromal TILs were scored in tumor samples at baseline (BL) and surgery (if residual disease, RD). TILs were evaluated as a continuous and binary variable using a threshold selected based on BL distribution. Fisher’s exact test was used to compare TILs between CT and CT+P, and between RCB groups. Association with pathologic complete response (pCR) and with recurrence-free (RFS), distant recurrence-free (DRFS) and overall survival (OS) were evaluated using logistic regression and Cox proportional hazards models, respectively, adjusting for clinical stage and neoadjuvant treatment. All statistical tests were two-sided (significance p < 0.05). Results: Between 5/2019-1/2024, 121 TNBC pts were identified; 18 (14.9%) stage I, 49 (40.5%) stage II, 52 (43.0%) stage III, 2 (1.7%) unknown. 19 (15.7%) had germline BRCA1/2 mutations. 49 (40.5%) received CT and 72 (59.5%) CT+P, with anthracycline-based CT in 40 (81.6%) and 62 (86.1%) pts, respectively. The pCR rate was 51.0% (25/49; 95% CI 36.3-65.6%) with CT alone and 58.3% (42/72; 95% CI 46.1-69.8%) with CT+P. Median follow-up was 2.2 years (interquartile range [IQR] 1.4-3.9): CT, 4.2 years (IQR 3.4-5.1); CT+P, 1.7 years (IQR 1.2-2.2). 115 pts had TILs scored at BL and 41 at RD, of whom 35 had paired samples. BL TILs (range 0-90%) were categorized by quartiles (5%, 10%, 30%) with 10% used for primary analysis. Both BL and RD TILs did not differ between CT vs CT+P as a continuous variable (BL p=0.11; RD p=0.56) or ≥10% cutoff (BL p=0.62; RD p=0.56). High BL TILs (≥ 10%) were significantly associated with pCR and this did not differ by treatment (interaction p=0.36). No association was observed between TILs (at BL or RD) and RCB class (TILs as continuous variable: BL p=0.36, RD p=0.84; or 10% cutoff: BL p=0.31, RD p=0.86). High BL TILs (≥ 10%) were associated with longer RFS (aHR 0.15, p=0.01), DRFS (aHR 0.17, p=0.01) and OS (aHR 0.19, p=0.05; or as a continuous variable, aHR per 5-unit increase 0.46, p=0.05). Neither TILs in RD nor increase from BL to RD (in 14/35 pts, 40%) were associated with RFS, DRFS or OS. Conclusion: In this multicenter TNBC cohort, BL TILs were reinforced as a prognostic marker for pCR, regardless of the addition of P to CT. TILs at RD and changes during neoadjuvant therapy were not prognostic. Additional follow-up for survival outcomes is ongoing. Citation Format: R. S. Lee, Q. Jin, B. Binboga Kurt, B. Koca, J. Gomez Tejeda Zanudo, A. Patel, A. Barkell, J. Baginska, O. M. Cunningham, C. E. Stever, T. S. Parker, T. Rahman, M. Luo, I. G. Martino, B. M. Drummey, S. A. Virani, K. Santos, J. Bsat, C. Snow, N. Tung, S. Lo, M. G. Faggen, N. Sinclair, N. Ahmad, M. Constantinou, S. Sinclair, J. L. Meisel, S. A. Kirschner, T. A. King, R. Salgado, E. A. Mittendorf, N. U. Lin, N. Tayob, E. P. Winer, E. C. de Bruin, S. M. Tolaney, A. Moeini, A. C. Garrido-Castro. Association of tumor-infiltrating lymphocytes (TILs) with outcomes in patients with early TNBC treated with neoadjuvant chemotherapy with or without pembrolizumab [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS4-07-18.

Abstract Genomes have the capacity to drastically modify hypotheses about the relationships of species. Despite the growing availability of non-model organism genome sequences, historically contentious portions of Tree of Life remain untested using genomic data. Here, we infer the phylogeny of sharks, skates, rays, and chimaeras using the genomes of 48 species, targeting different genomic marker types. Although phylogenetic relationships of chondrichthyans are relatively consistent across analyses, different molecular markers yield conflicting results about shark monophyly. Exons support the traditional view that sharks are monophyletic, whereas ultraconserved elements and legacy nuclear markers instead suggest that the frilled and cow sharks ( Hexanchiformes ), which retain the ancestral jaw structure of cartilaginous fishes, is the sister lineage of all other sharks and rays. The resolution of sharks as monophyletic or paraphyletic has little effect on inferences of the timescale of shark evolution or the origins of key traits, such as their ancestral ecology and genome size. We tie the diversification of living cartilaginous fishes to the transformation of marine ecosystems during the middle Mesozoic Era, confirming that living shark diversity is the product of rapid ancient diversification. Consequently, our results suggest that despite uncertainty around whether sharks are monophyletic, consensus can still be reached about major evolutionary events in this iconic vertebrate lineage. Significance Statement Living sharks, skates, rays, and chimaeras form one of the three principal groups of vertebrates. These iconic animals, which include over 1200 species, are key components of marine ecosystems and have helped us reconstruct the evolution of vertebrate genomes and phenotypes. However, much work on this group has assumed that sharks are a natural group. Here, for the first time, we leverage genome-scale data to test this hypothesis. Surprisingly, we show different genome regions reject or support hypothesis that sharks form a natural group to the exclusion of skates and rays. This throws an unexpected wrench into our understanding of the relationships of some of the oldest living vertebrate clades.

Objective To propose a revised neurobiological framework for traumatic memory that explains the phenomenology of flashbacks not as mere reactivations of fear circuits, but as the hippocampal complex’s ongoing attempt to integrate unprocessed sensory–emotional fragments encoded without temporal context during extreme stress. Background Classical models—the fear-network theory (Foa & Kozak, 1986) and the dual-representation theory (Brewin et al., 1996, 2010)—conceptualize flashbacks as intrusive replays of amygdala-driven sensory representations that fail to link with hippocampal–prefrontal systems. These frameworks accurately describe fragmentation but overlook the hippocampus’s potential active role in later integration. Neurobiological and computational findings across the last two decades show that traumatic stress disrupts prefrontal inhibition, amygdala dominance, and hippocampal contextual encoding, resulting in memories stripped of temporal and narrative structure. The persistence of flashbacks despite safety suggests a dynamic process rather than a static lesion. Methods / Theoretical Approach This chapter synthesizes evidence from functional neuroimaging, neuroendocrinology, and predictive-coding theory to construct a tri-systemic model of trauma involving the medial prefrontal cortex, amygdala, and hippocampus. The analysis follows the chronological breakdown of this system under threat—prefrontal disinhibition, amygdala hyper-encoding, HPA-axis discontrol, and hippocampal silencing—and explores subsequent re-engagement mechanisms. Results / Proposed Mechanism Flashbacks are conceptualized as failed integration events: partial hippocampal reactivation of unbound sensory–emotional traces in the absence of temporal coding. This re-engagement triggers amygdala-mediated affective flooding, creating the illusion of present-tense re-living. Conclusions / Implications Re-experiencing may represent a neural system striving for coherence, not solely pathology. Understanding flashbacks as integration attempts clarifies why they can emerge during calm, why therapies restoring safety and context diminish them, and how hippocampal–prefrontal synchronization transforms traumatic immediacy into autobiographical memory. This model reframes trauma as a process of disrupted, yet potentially recoverable, neurobiological integration.

Big chatbots like ChatGPT and Gemini are very good at talking, but they often forget what you said before[1]. If you have a long chat or come back another day, it feels like the AI has no memory. This can be frustrating when you want the AI to remember what you already told it. In this paper, we share a new idea called smart indexing[2]. Think of it like giving the AI a notebook with an index, so both you and the AI can quickly find important things from past talks. Our system does three special things. First, it gives you a side index view, like the table of contents in a book. This makes it easy to jump back to a topic you talked about earlier. Second, it uses topic detection, so when the subject changes, the AI can group parts of the chat into chapters and subtopics. This keeps the conversation neat and organized. Third, it can remember across different days with multi-session memory, so your chats do not disappear after each session. Together, these features make AI feel more helpful and human. It does not just talk, it remembers, organizes, and grows with you over time. This can be useful in school, hospitals, offices, and even research projects.