Citations

8021 Background: Lung cancer is the leading cause of cancer-related mortality worldwide, primarily due to delayed diagnosis and diagnostic variability in histopathological assessment. Advances in deep learning applied to medical imaging offer the potential to enhance early detection accuracy, reduce interobserver variability, and improve diagnostic efficiency in lung cancer. Methods: In this retrospective study, we evaluated DAVINCI using a dataset of 8,962 patients with confirmed lung malignancies. The model utilizes a specialized architecture combining convolutional neural networks (CNNs) and residual blocks to optimize feature extraction from high-resolution pathological slides. Performance was measured via detection accuracy, recall, specificity, and Area Under the Curve (AUC). Results: A total of 8,962 patients were included in the analysis. The DAVINCI model correctly identified lung cancer in 93% of cases, corresponding to 8,334 of 8,962 patients, with a 95% confidence interval (CI) of 86%–100% (7,707–8,962 patients). Model specificity was also 93% (95% CI, 86%–100%), indicating a high true-negative rate. Sensitivity (recall) was 87%, with correct detection in 7,797 patients and a 95% CI of 80%–94% (7,170–8,425 patients). The model demonstrated strong discriminatory performance, achieving an area under the curve (AUC) of 0.91 (95% CI, 0.84–0.98). In comparison, active pathologists achieved a mean diagnostic accuracy of 83%, corresponding to 7,438 of 8,962 cases, with a 95% CI of 76%–90% (6,811–8,066 cases). In terms of efficiency, DAVINCI processed whole-slide images in 16–19 seconds, substantially faster than conventional manual review, which typically requires several minutes per slide. Conclusions: DAVINCI demonstrated superior accuracy, robust discriminatory performance, and markedly faster processing compared with human review, supporting its potential role in early lung cancer detection at scale.

8000 Background: The NCI’s ALCHEMIST program has screened thousands of patients with resected lung adenocarcinoma for mutations in EGFR and ALK. EA5142 studied the efficacy of adjuvant nivolumab after standard of care adjuvant therapy in patients with resected lung adenocarcinoma without sensitizing EGFR and ALK alterations and squamous cell carcinoma. Methods: Patients enrolled in the ALCHEMIST screening trial (A151216) with resected NSCLC tumors at least 4 cm and/or lymph node positive (N1/2) (for lung adenocarcinoma, without sensitizing EGFR and ALK alterations), were centrally tested for tumor cell PD-L1 expression using the DAKO 28-8 clone. After completion of all planned adjuvant therapy (chemotherapy and/or post-operative radiation) patients were randomized 1:1 to adjuvant nivolumab 480 mg IV every 4 weeks for up to 1 year or best supportive care. Patients were stratified by tumor histology, stage (AJCC 7 th edition IB/IIA vs IIB/IIIA), prior adjuvant treatment, and PD-L1 status (<1% vs ≥1%). The primary study endpoints were disease free survival (DFS) in all patients and in patients with high tumor PD-L1 expression (≥50%). Overall survival was powered to be tested hierarchically, if DFS was positive. Results: Between July 2016 and October 2019, 935 patients were randomized across 378 sites. The treatment arms were balanced for age (median 66 years), sex (52% male), smoking status (10% never smoked), histology (29% squamous), stage and PD-L1 expression (29% PD-L1 high). With a median follow-up of 72.6 months, nivolumab did not improve DFS in the intention to treat population (median DFS 71.3 months vs. 68.8 months in observation arm), hazard ratio (HR) 0.97 (95% CI 0.81-1.17, p=0.78) or the PD-L1 ≥50% subset (median DFS 89.8 months vs. 78.5 months in observation arm), HR 0.86 (95% CI 0.59-1.25, p=0.43). In multivariate analysis, adjusting for age, gender, and smoking history, the DFS HR was 0.99 (95% CI 0.82-1.19, p=0.89) in all randomized patients and 0.82 (95% CI 0.56-1.2, p=0.31) in the PD-L1 ≥50% subset. As the primary recurrence endpoint included any lung cancer, including new primaries, a sensitivity analysis was performed excluding new primary lung cancers, DFS HR 0.97 (95% CI 0.80-1.18, p=0.76) in all patients and HR 0.88 (95% CI 0.60-1.30, p=0.52 in the ≥50% subset. The median duration of treatment with adjuvant nivolumab was 9.4 months (range 0-12.7 months). The most common reasons for nivolumab discontinuation were adverse event (29%), patient withdrawal (11%), and disease progression while on treatment (10%). Treatment related grade 3-5 toxicities were reported in 25% of patients: Grade 3: 23%, Grade 4: 2%, Grade 5: <1%. Conclusions: In patients with resected NSCLC (≥4cm or LN+, EGFR/ALK -) adjuvant nivolumab did not improve DFS, irrespective of tumor PD-L1 expression. Clinical trial information: NCT02595944 .

7563 Background: Subcutaneous (SC) drug delivery options have become more widely available in multiple myeloma (MM). SC delivery of isatuximab (Isa), an anti-CD38 monoclonal antibody, has been investigated across clinical trials via a novel hands-free on-body injector (OBI). Isa SC OBI features a small, retractable needle, establishing the first device-enabled delivery of an oncologic treatment. The Phase 3 IRAKLIA, Phase 2 IZALCO, and Phase 1b (TCD15484) trials consistently demonstrated similar efficacy and safety of Isa SC OBI vs Isa IV in relapsed/refractory MM (RRMM) patients (pts). Here, we present a cross-trial analysis exploring safety and reliability of Isa SC OBI in RRMM pts. Methods: Pts with RRMM in IRAKLIA (N=263) and the Phase 1b expansion cohort (N=22) received 1400 mg Isa SC OBI plus pomalidomide and dexamethasone (d). In IZALCO, RRMM pts (N=64) received 1400 mg Isa SC OBI + carfilzomib (K)-d. Pts were monitored for 4 hrs (IRAKLIA) or 1-2 hrs (IZALCO) following first administration and for 1 hr for the following cycle (C) 1 administrations to assess local tolerability. Pre-infusion medications included montelukast (C1), steroids, acetaminophen and H1 antihistamines. Pts without systemic infusion-related reactions (IRRs) after 4 consecutive administrations of Isa SC OBI had subsequent pre-medication reassessed at investigator’s discretion. Safety and device performance explored in this cross-trial analysis of company sponsored studies were assessed in the Isa SC OBI cohorts by the duration of injection, IRRs, local injection-site reactions (ISRs), and injection/device success, which were defined similarly across all trials. Results: The median duration of Isa SC OBI injection was 13 mins in IRAKLIA, 12 mins in IZALCO, and 10 mins in the Phase 1b trial (6426 total injections, N=349 pts). IRRs across trials occurred in 6/6426 (0.09%) injections and 4/349 (1.15%) pts. IRRs were mostly grade 1 (G; 3/6426, 0.05%) with 1 G3 (0.02%); none leading to discontinuation. In IRAKLIA, 80/263 (30.4%) pts did not require premedication (931/5145 injections, 18.1%); no IRRs occurred in this group. Across trials, ISRs occurred in 21/349 (6.02%) pts resulting in 35 (0.54%) ISR events. Most ISRs were G1 (33, 0.51%), the remaining were G2 (2, 0.03%) and generally occurred on the day of injection. 99.9% (6421/6426) of injections in 98.6% (344/349) pts were successful (completed without interruption) across all trials. Conclusions: These findings support overall low, self-limiting rates of IRRs and ISRs across trials, with >6000 (99.9%) Isa SC OBI injections successfully delivered. Isa SC OBI shows reproducible safety and reliability in IRAKLIA, IZALCO and Phase 1b trials, supporting it as a novel administration option for MM pts and its potential for future exploration in at-home administration. Clinical trial information: NCT04045795 , NCT05405166 , and NCT05704049 .

7514 Background: Iberdomide (Iber) is a CELMoD that binds with greater specificity & affinity to the cereblon protein of the E3 ligase complex than a traditional IMiD. Both transplant-eligible (TE) & ineligible (TIE) patients (pts) with newly diagnosed multiple myeloma (NDMM) are currently treated with quadruplet induction, consisting of an anti-CD38 monoclonal antibody, a proteasome inhibitor, an IMiD & dexamethasone (dex). We conducted a clinical trial (NCT05392946, IDEAL) to examine the efficacy of a novel finite-duration approach, using Iber, daratumumab, bortezomib & dex (Iber-DVd) quadruplet induction followed by Iber monotherapy maintenance. Methods: Both TE and TIE pts with NDMM were enrolled in this Phase 1 (3+3 design)/Phase 2 (single stage) trial. The primary goal was to assess the ≥complete response (CR) rate — utilizing the more sensitive serum MASS-FIX assay in lieu of serum immunofixation — during induction. Treatment involved twelve 28-day cycles of induction with Iber, given PO at the recommended phase 2 dose (RP2D), days 1-21, daratumumab 1800mg SQ, weekly for 2 cycles, every other week during Cycles 3-6, & every 4 weeks thereafter, bortezomib 1.3 mg/m 2 SQ, weekly & dex 40 mg, weekly, followed by Iber monotherapy (24 cycles). Growth factor was not allowed during the dose-limiting toxicity (DLT) assessment period (Cycle 1). Results: Among 47 pts enrolled, 9 pts were in Phase 1 portion which assessed the maximum tolerated dose (MTD), with Iber starting at 1 mg/day dose, days 1-21 [Dose Level (DL) 1]. DLT (all grade 4 neutropenia) was noted in 3 pts, 2 at DL 1 & 1 at DL -1. This analysis includes 44 pts who received Iber at the RP2D (0.75 mg/day, days 1-21), including 6 pts from the Dose Confirmation Cohort (Phase 1, DL -1), dosed at MTD, & 38 pts in the Dose Expansion Cohort. The median age at registration was 65 years (range 36-86 years), & 23 pts (52.3%) were deemed high-risk. At data cut-off, pts had received a median of 15 cycles (range 3-36) of therapy. Overall response rate was 100%; ≥CR rate, 36.4% (95% CI: 22.4, 52.2) during induction & 52.3% (95% CI: 36.7, 67.5), overall, including the maintenance phase. The marrow flow-based (10 -5 ) MRD negative rate was 29.5% during induction & 47.7% overall. At a median follow-up (FU) 18.3 mos, 2 pts have died (one during FU due to progressive disease & the other on-treatment due to COVID-19). 12- (& 18-) month progression-free survival & overall survival rates were 91% (88%) & 97% (94%), respectively. Most common toxicities at RP2D were lymphopenia, neutropenia (leading 27% of pts to require pegfilgrastim support), rash, peripheral neuropathy, diarrhea, & infections. Cases of high-grade fatigue were low (<5%). Updated data will be presented at the meeting. Conclusions: Iber-DVd is an active and safe regimen for both TE and TIE pts with NDMM, leading to high response rates which improved over time, with nearly one-half of the pts achieving MRD negative state. Clinical trial information: NCT05392946 .

7041 Background: CD20×CD3 bispecific antibodies and CD19-directed CAR T-cell therapies are increasingly used in relapsed/refractory DLBCL. CAR T therapy is associated with higher rates of CRS,ICANS, and cytopenias, whereas bispecific antibodies are often perceived as less toxic and more outpatient-friendly. Methods: We conducted a retrospective cohort study using the TriNetX Global Collaborative Network. Adult patients with DLBCL treated with bispecific antibodies (glofitamab, epcoritamab) were compared with those receiving commercial CAR T-cell therapy (tisa, liso, axi). Propensity score matching was performed on demographics, comorbidities, baseline laboratory values, and corticosteroid exposure, yielding 468 patients per cohort. Outcomes included CRS, ICANS, severe neutropenia (ANC <500/µL), and thrombocytopenia (platelets <50×10³/µL), as well as hospitalization and ICU admission at 1, 3, 6, and 12 months. Overall survival (OS) was assessed using Kaplan–Meier analysis with log-rank testing. Results: After matching, CAR T therapy was associated with significantly higher immune-mediated toxicity. CRS occurred more frequently with CAR T at 1 month (37.4% vs 22.4%), 3 months (39.3% vs 27.6%), 6 months (40.6% vs 28.8%), and 12 months (41.2% vs 29.7%) (all p<0.001). ICANS was also more common with CAR T across all timepoints (1 month: 10.5% vs 4.9%; 12 months: 12.6% vs 7.5%). Severe neutropenia and thrombocytopenia were markedly more frequent in the CAR T cohort, with neutropenia affecting 66.7% vs 12.8% at 1 month and 71.6% vs 32.7% at 12 months (all p<0.001). Despite higher toxicity, healthcare utilization was comparable. Hospitalization rates were similar between groups at 1 and 3 months (61.8% vs 66.7% and 67.1% vs 75.9%), and ICU admission rates did not differ meaningfully at any timepoint. Importantly, overall survival consistently favored CAR T therapy. Survival differences emerged early and persisted over time, with lower mortality in the CAR T cohort at 1 month (2.1% vs 9.0%), 3 months (7.7% vs 18.6%), 6 months (15.8% vs 26.7%), and 12 months (26.5% vs 35.3%), corresponding to significantly improved OS by log-rank testing at all timepoints (p<0.001). Conclusions: In this large, real-world, propensity-matched analysis of relapsed/refractory DLBCL, CAR T-cell therapy was associated with substantially higher rates of CRS, ICANS, and severe cytopenias compared with bispecific antibodies. However, these toxicities did not translate into higher hospitalization or ICU utilization. Despite a less favorable early toxicity profile, CAR T therapy conferred a consistent and clinically meaningful survival advantage from early to late follow-up. Early toxicity alone should not preclude CAR T referral in appropriate candidates, underscoring the need for optimized patient selection, sequencing strategies, and supportive care.

7011 Background: Up to 40% of pts with diffuse large B-cell lymphoma (DLBCL) will relapse or have refractory disease following 1L treatment (Tx) with SOC chemoimmunotherapy regimens R-CHOP or Pola-RCHP. High risk (HR) populations include those with International Prognostic Index (IPI) score 3–5 and IPI 1–2 with either ≥1 lesion ≥7 cm or baseline LDH >1.3×ULN. GOLCA is a potential, first-in-class, oral CELMoD designed for the Tx of lymphoma, with preferential distribution to lymphoid organs and enhanced activity in lymphoma cell lines. GOLCA induces rapid and deep degradation of Ikaros and Aiolos, leading to direct cell killing and immunomodulatory activity. GOLCA 0.4 mg + R-CHOP has shown promising safety, a ≥90% minimal residual disease (MRD) negativity rate, and durable efficacy (24-mo PFS, 79%) in 1L a-BCL (Nowakowski et al, ASH 2025, 476) and is being evaluated in the Phase (Ph) 3 study GOLSEEK-1 (NCT06356129). We report initial Ph1b results with GOLCA + Pola-RCHP in 1L a-BCL. Methods: Pts aged ≥18 y with ECOG PS ≤2, IPI 0–5 (Part 2A) or 2–5 (Part 2B), and newly diagnosed a-BCL with measurable tumorous disease were included. Pts were treated with GOLCA + Pola-RCHP for six 21-day (D) cycles or until disease progression/unacceptable toxicity/study withdrawal. In dose escalation, GOLCA was dosed at 0.2 and 0.4 mg D1–7. In dose expansion, pts were randomized 1:1 to Pola-RCHP + GOLCA 0.2 or 0.4 mg. Primary endpoint: safety; efficacy was a secondary endpoint. Results: At data cutoff, 57 pts were treated with GOLCA + Pola-RCHP (0.2 mg, n=28; 0.4 mg, n=29). Median age was 63 y. Most pts had HR disease (86%) and DLBCL not otherwise specified histology (84%). For known cell of origin (COO), 47% were germinal center B cell (GCB) and 39% non-GCB by Hans algorithm. Forty-seven pts (83%) completed Tx and 52 (91%) entered the follow-up (f/u) period. Grade 3/4 Tx-emergent adverse events occurred in 95% of pts, most frequently neutropenia (86%; an on-target side effect of GOLCA) and anemia (39%). Median relative dose intensity for GOLCA and Pola-RCHP components was >95%. At 15.5 mo median f/u, end of Tx (EOT) complete metabolic response (CMR) rate was 69% at 0.2 mg, 82% at 0.4 mg, and similar across COO subtypes. EOT CMR was consistent in HR pts (65% at 0.2 mg, 83% at 0.4 mg). The 12-mo DOR and PFS rates were 92% and 88% at 0.2 mg, and 96% and 96% at 0.4 mg. In pts with both pre-Tx and EOT ctDNA analysis, 65% and 83% at 0.2 and 0.4 mg achieved MRD negativity (PhasED-Seq) at EOT. Conclusions: GOLCA + Pola-RCHP demonstrated a predictable and manageable safety profile, and addition of GOLCA did not compromise median dose intensity of Pola-RCHP. GOLCA 0.4 mg + Pola-RCHP resulted in a high rate of durable CMRs, high rate of MRD negativity, and a promising 12-mo PFS rate. These data continue to show that GOLCA + SOC regimens R-CHOP and Pola-RCHP has potential to improve outcomes in 1L HR a-BCL. Clinical trial information: NCT04884035 .

632 Background: Prospective data on the risk of breast cancer (BC) recurrence among young BC survivors who have a subsequent pregnancy are limited. We sought to evaluate the long-term impact of pregnancy and live birth on BC outcomes in the Young Women’s Breast Cancer Study (NCT01468246), a prospective multicenter study of women aged ≤40 years at BC diagnosis. Methods: Women with stage 0-III BC without prior hysterectomy were included. The primary endpoint was breast cancer-free interval (BCFI) between patients with and without a (a) pregnancy and (b) live birth after BC. Secondary endpoints were distant recurrence-free interval (DRFI) and overall survival (OS). A time-varying Cox proportional hazards model was performed by estrogen receptor (ER) status, controlling for age at diagnosis, tumor stage, HER2 status, tumor grade, parity at diagnosis, and germline pathogenic variant status. Results: Among 1,016 BC survivors at a median follow-up of 12 (range, 0.5-18.5) years, 198 reported ≥1 pregnancy and 165 reported ≥1 live birth post-diagnosis. Among survivors who reported a post-diagnosis pregnancy, median age at diagnosis was 32 (range, 17-40) years; most had stage I (35%) or II (40%) BC, with 72% hormone receptor (HR)-positive and 25% HER2+; 42% were nulligravid and 60% were nulliparous at diagnosis. Among BC survivors who did not become pregnant post-diagnosis, median age at diagnosis was 37 (range, 21-40) years; most had stage I (35%) or II (43%) BC, with 75% HR-positive and 28% HER2+; 25% were nulligravid and 29% were nulliparous at diagnosis. Among BC survivors with ER-positive tumors (n = 737), pregnancy after BC was not associated with a difference in BCFI (adjusted hazard ratio [HR] 0.60, 95% CI 0.31-1.16, P = 0.130), DRFI (HR 0.73, 95% CI 0.35-1.55, P = 0.416), or OS (HR 0.58, 95% CI 0.24-1.40, P = 0.227). Similarly, live birth after BC did not impact BCFI (HR 0.83, 95% CI 0.45-1.55, P = 0.560), DRFI (HR 1.04, 95% CI 0.52-2.06, P = 0.923), or OS (HR 0.77, 95% CI 0.34-1.76, P = 0.541). Outcomes among BC survivors with ER-negative tumors (n = 278) were also not affected by post-diagnosis pregnancy (BCFI [HR 1.48, 95% CI 0.70-3.14, P = 0.304], DRFI [HR 1.59, 95% CI 0.68-3.75, P = 0.287], OS [HR 0.99, 95% CI 0.39-2.49, P = 0.975]) or live birth (BCFI [HR 1.02, 95% CI 0.41-2.53, P = 0.974], DRFI [HR 0.80, 95% CI 0.24-2.69, P = 0.722], OS [HR 0.39, 95% CI 0.09-1.69, P = 0.208]). Conclusions: In this multicenter, prospective study with 12 years of median follow-up, pregnancy and live birth after BC did not impact BC events or overall survival, irrespective of ER status. These long-term data from a modern cohort provide reassurance for young BC patients interested in future fertility. Clinical trial information: NCT01468246 .

6501 Background: Hepcidin, a central regulator of iron homeostasis, is pathologically elevated in patients with myelofibrosis (MF) and anemia. Chronic hepcidin elevation limits iron availability for red blood cell (RBC) production and contributes to anemia onset and severity. DISC-0974 is an investigational, first-in-class, monoclonal antibody that blocks hemojuvelin, a co-receptor in the bone morphogenetic protein-signaling pathway driving hepcidin expression. Methods: RALLY-MF (NCT05320198) is an ongoing Phase 2, open-label study conducted in the US. The primary aim is to evaluate DISC-0974 efficacy in participants with MF and anemia. Secondary aims include evaluating safety, pharmacokinetics (PK), and pharmacodynamics (PD). Eligible participants are ≥18 years of age with primary or secondary MF and hemoglobin (Hgb) <10 g/dL or RBC transfusion requirement. Participants enroll into 3 cohorts based on RBC units transfused in the 84 days prior to screening: non-transfusion dependent (nTD, 0 units and Hgb <10 g/dL), low transfusion burden (TD low, 1-2 units), high transfusion burden (TD high, 3-12 units). Stable dose of concomitant Janus kinase inhibitor (JAKi) is allowed. DISC-0974 is administered subcutaneously at 50 mg monthly for up to 6 doses, with escalation to 75 mg for inadequate or lost response. Primary endpoints include major hematologic response defined as transfusion independence (TI) during any consecutive 16 weeks (TD low) or 12 weeks (TD high), or a mean Hgb increase of ≥1.5 g/dL from baseline for ≥12 weeks (nTD). Secondary endpoints include PK/PD markers of iron regulation and safety assessments. Data were summarized using descriptive statistics. Results: At data cut (16 Oct 2025), 47 participants enrolled in the nTD (n=30), TD low (n=10), and TD high (n=7) cohorts, with 53% on concomitant JAKi therapy. DISC-0974 led to sustained hepcidin reduction and iron mobilization across cohorts. DISC-0974 resulted in meaningful hematologic responses among evaluable participants: 50% of nTD participants achieved a mean Hgb increase of ≥1.5 g/dL for ≥12 weeks, 71% of TD low participants achieved TI for ≥16 weeks, and 67% of TD high participants experienced a ≥50% reduction in transfusion burden, with follow-up ongoing. Major hematologic response rates were 50% for participants on and off concomitant JAKi therapy (n=18 and n=16, respectively). DISC-0974 was associated with meaningful improvement in FACIT-Fatigue scores for nTD and TD low participants. Serious adverse events (AEs, n=9) and ≥Grade 3 AEs (n=16) were considered unrelated to DISC-0974. There were no early withdrawals due to AEs. Conclusions: DISC-0974 was well tolerated and shows a favorable safety profile. Anemia response and fatigue scores improved across patients with MF and anemia, including those receiving JAKi therapy. These data validate hepcidin reduction as a promising approach to treating anemia in MF. Clinical trial information: NCT05320198 .

603 Background: Accurate prediction of pathologic complete response (pCR) after preoperative therapy for HER2-positive (HER2+) breast cancer can help triage patients to immediate surgery vs additional preoperative therapy. Pre-surgery (i.e., after neoadjuvant therapy) imaging studies may help estimate the likelihood of pCR. In EA1181, patients (pts) with stage II-IIIA HER2+ breast cancer received four cycles of preoperative taxane, trastuzumab, and pertuzumab (THP×4). Pre-surgery imaging was required, though the specific imaging modality was not mandated. Here, we report the accuracy of pre-surgery breast MRI to predict pCR (ypT0/Tis, ypN0). Methods: Among 2,141 pts treated with THPx4, 1,378 had pre-surgery breast MRI and were representative of the overall cohort, although T3 tumors were more common (16% vs 9%). MRI reports were available for 1,351 pts; all reports were centrally reviewed and classified as radiologic complete response (rCR; no residual tumor enhancement and normalization of lymph nodes, if applicable) or radiologic residual disease (rRD). Reports with indeterminate radiologic response were independently reviewed by two study radiologists blinded to pathology outcomes. Among pts with rCR who had pathologic residual disease (pRD), we also assessed the extent of residual disease by residual cancer burden (RCB) score. Among pts with rRD who had a pCR, we determined the proportion who only had residual ductal carcinoma in situ (DCIS). Results: See Table. Conclusions: After neoadjuvant therapy, preoperative MRI accurately predicted pCR in 86% of patients with HER2+/ER− disease and may help identify patients who can proceed directly to surgery. MRI is insufficient to predict pCR for pts with HER2+/ER+ disease, as only 55% with complete response on MRI had a pCR at surgery. A pre-surgery MRI that indicated residual disease was accurate in 80% of pts with HER2+/ER+ disease, but in only 45% of pts with HER2+/ER- cancer. Accurately predicting residual disease after THPx4 allows for consideration of additional therapy that may increase the chance of pCR. Combining imaging like MRI with other biomarkers, such as tissue-based markers or circulating tumor DNA, may improve prediction of pCR and facilitate personalization of treatment decisions in HER2+ breast cancer. All pts with pre-surgery MRI (n=1351)pCR rate overall in this cohort = 47% MRI result Pathology pCR pRD rCR 70.6% (341/483) 29.4%(142/483) 78% had RCB1 rRD 34%(295/868) 21% of 295 had DCIS only 66.01% (573/868) ER- (n=589)*Overall pCR rate in ER- subset = 68% MRI result Pathology pCR pRD rCR 86.1% (205/238) 13.9% (33/238) 85% had RCB1 rRD 55.3% (194/351) 19% of 194 had only DCIS 44.7% (157/351) ER+ (n= 762)*Overall pCR rate in ER+ subset = 31% MRI result Pathology pCR pRD rCR 55.5% (136/245) 44.5% (109/245) 76% had RCB1 rRD 19.5% (101/517) 26% of 101 had only DCIS 80.5% (416/517)

560 Background: Patients (pts) with early stage hormone receptor-positive, human epidermal growth factor receptor 2-negative (HR+/HER2-) breast cancer (BC) remain at both short and long term risk of recurrence. While clinico-pathological and gene-based biomarkers provide insight into early and late recurrence risk, additional prognostic tools are needed. We previously developed (Saha et al. Clin Cancer Res , 2025) a new breast MRI image-based artificial intelligence prognostic tool (TumorSight Risk) composed of both radiographic and clinical information. Here we sought to validate this tool in an independent cohort. Methods: In this multi-center, blinded, prospective/retrospective validation study, we sought to validate the 5-year recurrence-free survival (RFS) according to pre-defined risk scores (0-100) and categories (“high” or “low”) via pre-specified cutoff (≥47) for pts with HR+/HER2- BC. Kaplan-Meier and Cox regression analyses were then used to assess the unadjusted and adjusted association between risk category/score and the 5-year RFS. Results: 2,129 pts with HR+/HER2- BC diagnosed between January 1, 2010, and December 31, 2024, who underwent breast MRI prior to surgery (no neoadjuvant therapy) were included from four institutions. Median age was 57 years, 16% were Black, 15% had grade 3 disease, 93% were clinical T1 or T2, and 85%/15% were clinical N0/N1. At a median follow up of 6.24 years, the unadjusted hazard ratio (HR) for RFS comparing low (n=1808) and high risk (n=321) pts was 2.71 (95% CI = 1.95 – 3.76, p = 3x10 -9 ). The estimated 5 year RFS rate according to risk score (low and high) was 95.5% (95% CI: 94.63 – 96.3%) and 89.6% (95% CI = 86.4 – 92.6%) while the 10 year RFS estimates were 88.9% (95% CI = 86.5 - 90.9%) and 69.6% (95% CI = 59.1 – 77.9%). Univariate subgroup analysis demonstrated that the risk score was prognostic by age (< 50 vs. ≥50 years), pathologic nodal status, grade, and surgery type, regardless of chemo-, endocrine, or radiation therapy administration. The observed continuous risk score was strongly associated with risk of recurrence, with a 22% increase in relative risk per 10-unit change in risk score (0-100), translating to a 2.4% (95% CI = 1.2 – 4.5%) and 25.5% (95% CI = 15.4 – 40.4%) risk of recurrence at 5-years for pts at the lowest (0-9) versus highest (90-100) risk score. The prognostic association between risk score and RFS remained after adjusting for pathologic T- and N-stage, grade, age, race, and treatment (p = 0.002). Conclusions: TumorSight Risk is a new image-based biomarker for HR+/HER2- BC that provides prognostic information independent of standard clinico-pathological parameters. Given the ready availability of breast MRI in most pts with BC, TumorSight Risk may be a convenient tool for estimating risk for recurrence.