Citations

e19521 Background: Blindedindependent central review (BICR) committees are often established for oncology trials to assessbestoverall response (BOR), while real-world (RW) studies typically rely on physician-assessed responses. To emulate clinical trial rigor, a BICR committee was implemented in the R5458-ONC-21101 RW data-derived external control arm study (NCT05673967) to evaluate BOR using International Myeloma Working Group (IMWG) criteria. A pre-specified concordance analysis compared BOR assessments between the BICR committee and treating site physicians. Methods: RW data were obtained from chart reviews at 15 participating International Myeloma Foundation IMWG sites. The BICR committee, comprised of three multiple myeloma experts, evaluated BOR through consensus voting. Per protocol, the committee reviewed case report forms consisting of data on serum and urine myeloma protein and free light chain levels; and, where available, bone marrow plasma cell percentage, presence of plasmacytomas, and bone lesion changes. Concordance between BICR- and physician-assessed BOR was evaluated for objective response rate defined as ≥partial response (PR), ≥very good partial response (VGPR), ≥complete response (CR), and stringent complete response (sCR), with non-evaluable responses excluded from the analysis. Cohen’s kappa (κ) was used to measure agreement, with values >0.6 indicating strong concordance. 1 Results: Of 279 evaluable lines of therapy from 203 patients (Table), the BOR of ≥PR was 45.2% by BICR assessment and 42.3% by treating physicians (κ=0.622). Concordance improved for BOR of ≥VGPR (22.2% by BICR vs 24.7% by treating physicians; κ=0.731) but was low for BOR of ≥CR (3.9% vs. 10.8%; κ=0.508) and sCR (1.1% vs. 4.7%; κ=0.364). Conclusions: BICR- and physician-assessed responses showed strong concordance for BOR of ≥PR and ≥VGPR. However, as expected, lower concordance for ≥CR and sCR was observed, as evaluation of CR per IMWG criteria requires a bone marrow biopsy, which is rarely performed in clinical practice. While the reliability of CR and sCR in RW data is less certain, these results suggest that BOR of ≥VGPR or ≥PR may be reliably ascertained in academic settings based on treating physician evaluations. 1. Landis JR, Koch GG. Biometrics 1977;33(1):159–74. Clinical trial information: NCT05673967 . Agreement of BICR- and physician-assessed responses. BICR assessed Physician assessed Observed agreement, % Cohen’s kappa* Number of evaluable lines of therapy 279 279 - - ≥PR, n (%) 126 (45.2) 118 (42.3) 81.4 0.622 ≥VGPR, n (%) 62 (22.2) 69 (24.7) 90.3 0.731 ≥CR, n (%) 11 (3.9) 30 (10.8) 93.2 0.508 sCR, n (%) 3 (1.1) 13 (4.7) 96.4 0.364 *A kappa value >0.6 indicates strong concordance. 1

e19515 Background: The prevalence of diabetes mellitus (DM) is significantly higher in patients with multiple myeloma (MM) compared to the general population. DM has been associated with worse outcomes in MM, including reduced overall survival and increased mortality. The impact of diabetes in patients with MM is further compounded by cardiotoxicity from proteasome inhibitors and the increased thromboembolic risk linked to immunomodulatory agents. Given the potential benefits of sodium-glucose cotransporter-2 (SGLT2) inhibitors, we hypothesized that their use in MM patients with DM could improve clinical outcomes and mitigate adverse events in this population. Methods: We utilized the TriNetX network to identify patients (≥18 years old) with MM and DM between January 1, 2013, and January 1, 2019. We categorized the patients into SGLT2 inhibitors users and SGLT2 inhibitors users. Inclusion in the SGLT2 inhibitors group required exposure to SGLT2 inhibitors for at least 1 year. 1:1 Propensity score matching was performed to match these two groups based on comorbidities, medications, baseline hemoglobin A1c, low-density lipoprotein, and left ventricular ejection fraction to produce two comparable cohorts. The study outcomes include all-cause mortality, all-cause hospitalization, acute venous thromboembolism (VTE), acute heart failure (HF), acute myocardial infarction (MI), cardiac arrhythmia, and cardiac arrest over the 5-year follow-up duration. Results: A total of 43,078 patients were identified, including 3,640 SGLT2 inhibitors users and 39,438 non-SGLT-2 inhibitors users. After 1:1 propensity-score matching, there were 3,379 SGLT-2 inhibitor users (68.9 ± 10.3 years of age, 38.0% female, 23.9% African American [AA]) and 3,379 non-SGLT-2 inhibitor users (68.9 ± 11.0 years of age, 35.9% female, 22.2% AA). SGLT-2 inhibitors users were associated with lower rates of all-cause mortality (odds ratio [OR]: 0.34, 95% confidence interval [CI]: 0.30-0.39) and all-cause hospitalization (OR: 0.52, 95% CI: 0.47-0.58). Lower odds of acute VTE (OR: 0.59, 95% CI: 0.49-0.70), acute HF (OR: 0.89, 95% CI: 0.81-0.98), acute MI (OR: 0.81, 95% CI: 0.69-0.95), cardiac arrhythmia (OR: 0.82, 95% CI: 0.70-0.97), and cardiac arrest (OR: 0.57, 95% CI: 0.44-0.74) were reported among SGLT-2 inhibitors users compared to non-SGLT2 inhibitor users. Conclusions: Our study revealed that only 8.4% of patients with DM and MM were prescribed an SGLT-2 inhibitor. Our study also demonstrated that SGLT2 inhibitor use was associated with lower odds of all-cause mortality and hospitalization, reduced risk of VTE, and favorable cardiovascular outcomes in this patient population.

3624 Background: Increasing chemotherapy completion and reducing chemotherapy adverse events (AE) are critical to improve survival after colon cancer diagnosis. Sex, as a biological variable, may impact chemotherapy treatment differently, and thus further investigation is needed to examine sex differences in chemotherapy completion and adverse events among patients with colon cancer. Methods: Among an NCI-sponsored trial conducted among patients with stage III colon cancer (CALGB/SWOG 80702), all patients received standard adjuvant chemotherapy FOLFOX (fluorouracil, leucovorin, and oxaliplatin). To signal chemotherapy completion, we utilized relative dose intensity (RDI) calculated as the ratio of the delivered dose intensity to planned dose intensity; and reduced RDI (RDI <85%) was considered as a clinically significant deviation from standard FOLFOX. From clinicians’ records of NCI’s Common Terminology Criteria for Adverse Events (CTCAE), we primarily focused on clinically significant AE such as neutrophils decrease, nausea, platelets decrease, hypertension, peripheral neuropathy, diarrhea, fatigue, gastritis, creatinine increase, gastric ulcer, myocardial ischemia, and cerebral ischemia; and severe AE was defined as the occurrence of any above AE with CTCAE grade ≥3. Using multivariable logistic regression that adjusted for body surface area (BSA) and other clinicopathological confounders, we estimated adjusted odds ratios (OR) for the associations of sex with reduced RDI and severe AE. Results: Of 2201 patients, mean (standard deviation [SD]) age was 60.9 (10.9) years, 1019 (46.3%) were female, 1750 (79.5%) were White, 172 (7.8%) were Hispanic, 964 (43.8%) experienced reduced RDI, and 1156 (52.5%) had severe adverse events. Compared to males, females were at significantly higher risks of experiencing reduced RDI (OR [95% CI]: 1.57 [1.27-1.93], P <0.001) and severe AE (OR [95% CI]: 1.73 [1.41-2.12], P <0.001). Conclusions: Our findings suggested females are more likely than men to experience reduced RDI and severe AE during colon cancer chemotherapy. Clinical Impact: In the era of precision medicine, sex (as a biological variable) should be considered in optimizing colon cancer chemotherapy to improve completion and reduce toxicities. Clinical trial information: NCT01150045 .

1001 Background: Imlunestrant (imlu) is a next-generation, brain-penetrant, oral selective estrogen receptor degrader. The EMBER-3 trial, in patients (pts) with ER+, HER2- ABC who had disease progression on or after aromatase inhibitor-based therapy, showed significant progression free survival (PFS) improvement with imlu vs standard therapy (SOC, fulvestrant or exemestane) in pts with ESR1 mutations ( ESR1 m), and with imlunestrant+abemaciclib (imlu+abema) vs imlu in all pts, regardless of ESR1 m. Exploratory PRO analyses are presented here. Methods: EORTC QLQ-C30 was administered at baseline (BL) and every 8 weeks until treatment discontinuation. Prespecified QLQ-C30 analysis used a longitudinal mixed model for repeated measures to calculate mean change from BL in pts with BL and ≥1 post-BL score. PRO-CTCAE (diarrhea frequency) was administered weekly, reporting 0 (never) to 4 (almost constantly). PRO-CTCAE (injection site reaction [ISR]) was administered to fulvestrant recipients weekly for 2 weeks post-injection, reporting yes/no (pain, swelling, redness). Descriptive analysis was used for PRO-CTCAE. Results: In pts with ESR1 m, imlu monotherapy was associated with many improved or maintained EORTC QLQ-C30 scores, whereas scores with SOC were declined or maintained. Specifically, pts with ESR1 m on imlu had improved global health status (GHS)/quality of life (QOL) and physical function (PF) scores, while scores with SOC declined (mean change differences between treatments: 9.9 [0.1, 19.7] and 6.2 [-0.8, 13.1], respectively). These PRO findings mirror the PFS findings in this group. In the overall population, GHS/QOL scores declined similarly with imlu vs SOC (mean change differences: 0.5 [-4.7, 5.7]), while PF scores were maintained with imlu vs a slight decline with SOC (mean change difference: 2.5 [-1.1, 6.1]). Most fulvestrant recipients (72%) reported ISR at any time while on treatment, with a mean of 31% during the first week of the first 6 cycles. Imlu+abema vs imlu showed broadly similar declines in all pts, with minimal mean change differences in GHS/QOL and PF scores (0.8 [-7.4, 5.9]; -2.2 [-6.6, 2.2], respectively). Pts reported similarly low rates of “frequent” or “almost constant” diarrhea with imlu (3%) and SOC (2%) and higher rates with imlu+abema (22%). Conclusions: PROs from EMBER-3 demonstrated that patients with ESR1 m had better GHS/QOL and PF with imlu vs SOC, mirroring efficacy results. While the frequency of CTCAE defined ISRs was low, the high rate of PRO-CTCAE ISR demonstrates that this clinically relevant adverse event is underappreciated by physicians. Additionally, all pts had generally comparable GHS/QOL and PF with imlu+abema vs imlu. Overall, these results support the efficacy and safety of imlu compared to existing SOC. Clinical trial information: NCT04975308 .

e18161 Background: Thyroid cancer remains a significant cause of mortality globally, with varying trends across demographic, geographic, and socioeconomic groups. Understanding these patterns is critical for developing effective prevention and intervention strategies. This study examines thyroid cancer-related mortality trends in the United States from 1999 to 2023, focusing on age-adjusted mortality rates (AAMRs) and disparities across demographic and geographic classifications. Methods: This descriptive study utilized the CDC WONDER database to extract data on thyroid cancer (ICD-10 code C73) deaths among adults aged 25 and older from 1999 to 2023. AAMRs per 1,000,000 individuals were calculated and stratified by gender, race/ethnicity, census region, urban-rural classification, and state. Temporal trends were analyzed using Joinpoint regression to compute annual percent changes (APCs) with 95% confidence intervals (CIs). Statistical significance was set at p<0.05. Results: Between 1999 and 2023, there were 55,971 deaths attributed to thyroid cancer (Men: 24,398; Women: 31,573). Overall, the AAMR increased modestly, with an APC of 1.0 (95% CI: 0.8 to 1.2). Men exhibited a slightly higher AAMR increase (APC: 1.2; 95% CI: 1.1 to 1.4) compared to women (APC: 0.9; 95% CI: 0.7 to 1.2). Racial disparities revealed varied trends. Non- Hispanic White individuals experienced a steady AAMR increase (APC: 1.0; 95% CI: 0.8 to 1.2), while Hispanic populations showed a smaller increase (APC: 0.4; 95% CI: -0.1 to 1.1). Non-Hispanic Asians or Pacific Islanders had a slight decline (APC: -0.4; 95% CI: -1.1 to 0.5), and Non-Hispanic Black populations showed segmented trends, with an APC of 0.3 (95% CI: - 7.8 to 16.7) from 1999 to 2018 and a subsequent increase (APC: 3.7; 95% CI: -5.5 to 10.5) from 2018 to 2020. Geographically, the West exhibited the highest overall AAMR (11.0; 95% CI: 10.8–11.2), followed by the Midwest (9.9; 95% CI: 9.7–10.0) and Northeast (9.8; 95% CI: 9.6–10.0), with the South reporting the lowest AAMR (9.0; 95% CI: 8.8–9.1). Hawaii had the highest state-level AAMR (12.3; 95% CI: 10.8–13.7), while Georgia recorded the lowest (7.6; 95% CI: 7.2–8.1). Urban-rural disparities indicated a slightly higher APC in non-metropolitan areas (1.0; 95% CI: 0.7–1.4) compared to metropolitan areas (0.8; 95% CI: 0.6–1.0). Conclusions: Thyroid cancer-related mortality has shown a gradual increase in the United States over the past two decades, with distinct gender, racial, and geographic disparities. Men, non-metropolitan residents, and individuals in the West and Hawaii faced higher mortality burdens. These findings underscore the importance of targeted public health strategies to address these disparities and reduce thyroid cancer mortality in vulnerable populations.

e14601 Background: Immune checkpoint inhibitors (ICIs) provide a long-term survival benefit across multiple cancer types, however immune-related adverse events (irAEs) occasionally disrupt treatments and impact quality of life. Peripheral edema from ICIs can occur due to increased peripheral vascular permeability. Thus, we aimed to evaluate the incidence and outcomes of ICI-related peripheral edema. Methods: MEDLINE, Embase, and Web of Science were searched to identify phase 3 randomized controlled trials (RCTs) reporting peripheral edema events in patients treated with ICIs. Incidence rates of peripheral edema were analyzed based on treatment patterns (monotherapy/combination therapy) and ICI subtypes (PD-1/PD-L1/CTLA-4). A random effect model meta-analysis was utilized to calculate pooled odds ratios of edema incidence between treatment groups. Finally, a systematic review was conducted to summarize treatment outcomes of peripheral edema associated with ICIs. Results: Overall, 58 RCTs comprising 22,590 patients were identified for the incidence analysis. Peripheral edema as treatment-related adverse events (TRAE) (Grade 1-5) occurred in 2.80% of patients receiving ICI monotherapy (n = 195/6969; PD-1 inhibitors: 2.22% [n = 104/4694], PD-L1: 3.73% [n = 58/1557], CTLA-4: 4.60% [n = 33/718]), 4.96% (n = 112/2257) of patients receiving ICI with chemotherapy, and 8.73% (n = 205/2349) of patients receiving ICI with molecular-targeted therapy. Grade 3-5 TRAE peripheral edema occurred in 0.19% (n = 13/6969) of patients treated with ICI monotherapy. Meta-analysis showed ICI monotherapy had lower TRAE peripheral edema compared to chemotherapy (OR = 0.45, 95% CI [0.27, 0.73], p = 0.001). Subgroup analysis based on ICI subtypes showed high heterogeneity in subgroups (p , I 2 = 87.1%), but PD-1 inhibitors were associated with lower incidence than chemotherapy (OR = 0.41, 95% CI [0.23, 0.73], ). Subgroup analysis of chemotherapy regimens revealed that taxane chemotherapy was associated with a higher incidence of TRAE peripheral edema than ICI monotherapy (OR = 0.32, 95% CI [0.18, 0.58], p = < 0.001), but the edema incidence was comparable between ICI monotherapy and non-taxane chemotherapy (OR = 1.05, 95% CI [0.53, 2.08], p = 0.13). A systematic review of irAE peripheral edema treatment outcomes identified 19 studies comprising 40 patients. In total, 82.5% (n = 33/40) of cases received glucocorticoid treatment, with 55.0% (n = 22/40) requiring a high dose (≥ 1 mg/kg/day) of prednisone. ICI was discontinued in 77.5% of patients (n = 31/40) due to peripheral edema with permanent discontinuation in 35% (n = 14/40). Conclusions: Peripheral edema is an occasional but clinically significant side effect of ICI therapy. Management of irAE edema can result in permanent ICI discontinuation, which has implications for cancer treatment outcomes.

1098 Background: Leucine-rich repeat-containing protein 15 ( LRRC15) has emerged as a potential biomarker and therapeutic target for various cancers due to its high expression in cancer-associated fibroblasts (CAFs) and role in tumor progression. High LRRC15 expression is associated with poor prognosis in TNBC. This study aims to define the multiomic profile of LRRC15 in TNBC. Methods: 3,038 TNBC samples were analyzed via Next-Generation Sequencing (592, NextSeq; Whole Exome Sequencing, NovaSeq) and Whole Transcriptome Sequencing (NovaSeq; Caris Life Sciences, AZ). Immune cell fractions were estimated using WTS deconvolution (Quantiseq). Stromal cell abundance in the tumor microenvironment (TME) was estimated from RNA expression profiles using MCP Counter. LRRC15 -high (H) and -low (L) tumors were classified by RNA expression above or below the 25th percentile. Real-world overall survival (OS) and treatment-related survival were derived from insurance claims and calculated from tissue collection or treatment initiation to last contact using Kaplan-Meier. Statistical significance was assessed using chi-square and Mann-Whitney U tests with multiple comparison adjustments (q < .05). Results: LRRC15 -H TNBC tumors had higher frequency of PIK3CA (25.9% vs 16.8), PIK3R1 (6.2% vs 1.6%), PTEN (11.3% vs 5.8%), but lower frequency of RB1 (7.8% vs 12.1%) and KMT2D (2% vs 4.4%) compared to LRRC15 -L, all q < 0.05. LRRC15 -H had higher PD-L1 positivity (32.3% vs 24.5%, q < 0.05). Analysis of immune cells showed LRRC15 -H TNBC had higher infiltration of B cells (4.2% vs 3.5%), M1 macrophages (4.3% vs 2%), M2 macrophages (4% vs 2.5%), Tregs (1.9% vs 1.1%), neutrophils (4.9% vs 3.9%), CD8 + T cells (0.4% vs 0.1%), but lower dendritic cells (2.5% vs 3%), all q < 0.05. LRRC15 -H tumors had greater abundance of CAFs (575.6 vs 93.78, 6.14 fold change (FC)) and endothelial cells (7.3 vs 3.7, 1.97 FC), all q < 0.05. LRRC15 -H had higher T-cell inflamed score (71.5 vs -77) and IFNg score (-0.14 vs -1.72), all q < 0.05. LRRC15 -H tumors had higher expression of immune checkpoint genes ( CD274, PDCD1, PDCD1LG2, CTLA4, LAG3, HAVCR2, FOXP3, IDO1, TNFSF14, TIGIT, BTLA, CEACAM1, CD47, CD80, CD86, CD160, CD274 ; FC 1.2-2.5, q < 0.05). LRRC15 -H was associated with better OS (mOS: 24.7 vs 13.6 months; HR 0.61, 95% CI 0.53-0.7, p < 0.001). Post-pembrolizumab survival was longer for LRRC15 -H patients (mOS: 27.2 vs 19.4 months; HR 0.61, 95% CI 0.42-0.89, p = 0.01). Conclusions: LRRC15 -H TNBC exhibited better outcomes with pembrolizumab, likely due to higher immune cell fractions and increased CAFs. These findings highlight TNBC heterogeneity and position LRRC15 as a potential biomarker for tumor stratification, a possible adverse prognostic biomarker and a positive predictive biomarker. Ongoing phase I trials targeting LRRC15 show promise. Combining LRRC15 -targeted therapies with immunotherapy may improve TNBC outcomes, warranting further validation in breast cancer models.

1096 Background: Bria-IMT is a combination immunotherapy consisting of allogeneic whole cell cancer vaccine (SV-BR-1-GM) administered w/ immune checkpoint inhibitor (CPI). SV-BR-1-GM breast cancer cells are engineered to directly stimulate anti tumor immunity via expression of tumor associated antigens and secretion of GM-CSF to enhance dendritic cell activation. Addition of CPI potentiates SV-BR-1-GM to overcome the immune suppressive tumor microenvironment. Methods: This Ph I/randomized Ph II study evaluated the Bria-IMT regimen in pts w/ metastatic breast cancer; CTX (300 mg/m²) on day -2/-3, SV-BR-1-GM and CPI on Day 0, w/ low dose peg interferon α at inoculation sites on day 2 (±1) . Phase II pts were randomized 1:1 to receive CPI at cycle 1 or cycle 2. Two SV-BR-1-GM formulations (w/ vs w/o IFNγ incubation) were evaluated. Biomarkers included cancer-associated macrophage-like cells, circulating tumor cells, PD-L1 scores, and delayed-type hypersensitivity skin tests. Results: 54 pts (22 Ph I, 32 Ph II) enrolled; 11 received pembrolizumab, 44 retifanlimab (1 crossover). 33 (61%) pts were ER+/PR+/HER2-, 18 (33%) TNBC, 3 (6%) HER2+. Median OS, PFS, ORR, and CBR were evaluated against two pivotal Ph 3 trials, ASCENT 1 (SG in TNBC) and TROPiCS-02 2 (SG in HR+/HER2- MBC) (see Table 1). In randomized pts, C1 vs C2 CPI had PFS (3.7 vs 3.2 mos, P=0.09) and OS (11.4 vs 7.4 mos, P=0.19). Pts receiving Ph 3 formulation (w/o IFNγ; N=37) had greater PFS (3.6 vs 2.6 mos, P=0.01) and OS (13.4 vs 6.9 mos, P=0.01). Bria-IMT was well tolerated w/ no Tx related D/Cs. Conclusions: The Bria-IMT Ph 3 formulation cohort OS was comparable to ASCENT and TROPiCS-02 (13.43 vs 11.8, 14.4 mos), exceeding TPC arms (6.9, 11.2 mos). CBR (61%) compared favorably to ASCENT (40%) and TROPiCS-02 (34%); ORR (14%) matched or exceeded TPC arms (4%, 14%). These outcomes were observed in a more heavily pretreated population, demonstrating Bria-IMT’s clinical activity. Randomized Ph 2 results suggest efficacy and safety in heavily pretreated MBC, w/ no significant OS difference between C1 and C2 CPI initiation and 22% of pts still in active survival follow up. Superior outcomes w/ the Ph 3 formulation support its continued evaluation. A randomized Ph 3 trial is ongoing, comparing Bria-IMT vs treatment of physician’s choice (NCT06072612). Clinical trial information: NCT03328026 . Trial (Cohort) Age (Median, Range) Prior Therapies (Median) OS (Median, mos) PFS (Median, mos) ORR (%) CBR (%) Bria-IMT (Overall Cohort) 61 (38-81) 6 (2-13) 9.9 (1.8-30.3) 3.6 10% 55% Bria-IMT (Ph 3 Formulation) 62 (44-80) 6 (2-13) 13.43 (1.8-30.3) 3.6 (1.8-16.5) 14% 61% ASCENT (SG) 54 (27-82) 4 (2-17) 11.8 4.8 31% 40% ASCENT (TPC) 53 (27-81) 4 (2-14) 6.9 1.7 4% 8% TROPiCS-02 (SG) 57 (49-65) 3 14.4 5.5 21% 34% TROPiCS-02 (TPC) 55 (48-63) 3 11.2 4 14% 22% References: Bardia A et al. J Clin Oncol. 2024 May 20;42(15):1738-1744Rugo, Hope S et al. The Lancet, Volume 402, Issue 10411, 1423 – 1433.

1089 Background: T-DXd has become an important treatment option in mBC and other malignancies. Interstitial lung disease/pneumonitis (ILD) occurred in 10-14% of pts in the DESTINY-Breast trials (0-2% G5 ILD), and with potential lower incidence/severity in earlier line settings (Krop et al, ASCO 2023). We previously reported the Mayo Clinic Rochester, MN experience with T-DXd related ILD in mBC (Hoppenworth et al, ASCO 2024). Here, we expand the data to include patients treated at all locations of the Mayo Clinic Enterprise. Methods: We retrospectively identified pts with mBC who received ≥1 dose of T-DXd across the Mayo Clinic enterprise (Rochester, MN; Mayo Clinic Health System locations in MN/WI; Scottsdale, AZ; and Jacksonville, FL) between July 2022-December 2023. Demographic, mBC characteristics, and pulmonary clinical variables were abstracted from the clinical records. Data were summarized using descriptive statistics. Diagnosis of ILD was determined by treating clinicians, and severity was approximated to CTCAE V5 based on clinical documentation. Results: 252 pts with mBC received T-DXd during the study period. The majority were Caucasian (86%) and female (99%) with a median age of 63. 91 pts (36%) were current/formers smokers and 97 (39%) had prior pulmonary comorbidities. The majority had HER2 low (155, 62%) and HR positive (164, 65%) mBC. 35 (14%) developed any grade ILD [G1: 6 (17%), G2: 13 (37%), G3: 3 (8.6%), G4: 3 (8.6%), G5: 10 (29%)], with 15 (43%) presenting with ≥G3. 29 (83%) presented with at least 1 symptom (cough or SOB). Among those with previous pulmonary toxicity, 4 (33%) had previous pneumonitis in the ILD cohort. The median prior lines of all therapies (including endocrine therapy) were 5, with a median of 3 prior lines of chemotherapy in both the ILD (range 1-13) and non-ILD cohorts (range 1-13). Median onset to any grade ILD was 7 cycles. 20 pts (57%) received steroids for ILD. 14 (40%) had a bronchoscopy, all had a CT chest, 16 (46%) were hospitalized, 7 (20%) were intubated and 1 (3%) had a lung biopsy. Pts with G5 ILD had a median of 3 lines of chemo and a median of 8 cycles prior to onset of ILD, compared to 3 lines of chemo and a median of 7 cycles for those with G1-4 ILD. G5 pts were all Caucasian females, 5 were former smokers, 5 were non-smokers, 1 had a history of pneumonitis. T-DXd was rechallenged in 4 pts (G1-2 ILD) without ILD recurrence. Conclusions: In this retrospective case series,14% of pts treated with T-DXd experienced any grade ILD (4% G5) which is in alignment with the rate observed in the pivotal DESTINY-Breast trials, though with higher rates of G5 toxicity. Among those with ILD, increased lines of therapy were not associated with increased risk. Further research is needed to correlate risk.

e14626 Background: Despite advancements in neoadjuvant therapy and surgical techniques, patients with pancreatic ductal adenocarcinoma (PDAC) often experience high recurrence rates after total neoadjuvant therapy (TNT) and resection. Existing tumor response scoring systems, such as the College of American Pathologists (CAP) guidelines, fail to stratify recurrence risk effectively, emphasizing the need for novel, objective approaches. Methods: This study evaluated spatial characteristics of residual cancer and stroma in a multi-site cohort of 203 patients with resected PDAC following TNT, not having achieved a major pathologic response (CAP 2 or 3). Whole slide images (WSI) of hematoxylin and eosin-stained sections were digitized and analyzed using artificial intelligence to perform tissue segmentation. Metrics quantifying spatial morphology and configuration, inspired by principles of landscape ecology, were computed to explore spatial features of the tumor microenvironment (TME). Statistical models (recursive partitioning and regression tree [RPRT] model and Cox model with backwards eliminations) incorporating these features were developed to stratify recurrence risk. Results: Five metrics were used for risk classification model build. Three models successfully classified patients into high- and low-risk groups based on spatial metrics. Models demonstrated moderate discriminatory power (C-statistics: 0.560–0.566) and showed that increased tumor fragmentation, like stromal patch density and cancer patch shape index, were strongly associated with improved outcomes. Conclusions: Integrating spatial metrics of cancer-stroma configuration enhances risk stratification beyond conventional scoring systems. Given increased tumor fragmentation is associated with improved outcomes, these data indirectly suggest ecological mechanisms of treatment resistance. This approach offers a pathway toward personalized post-operative management strategies to improve outcomes in PDAC. Risk classification models. Model Description Event/n Median DFS (95% CI) HR(95% CI) p-value C-statistics 1(RPRT) High risk: log standard deviation of the stroma shape index > 0.651 87 / 101 8.22(5.72 – 10.1) 1.64(1.20 – 2.25) 0.002 0.562 Low risk: log standard deviation of the stroma shape index ≤ 0.651 71 / 102 11.90(9.24 – 25.1) Ref 2(RPRT) High risk: log standard deviation of the stroma shape index > 0.651 & mean cancer shape index ≤ 1.775 58 / 63 7.23(5.19 – 9.63) 1.82(1.31 – 2.52) 0.0003 0.560 Low risk: (log standard deviation of the stroma shape index > 0.651 & mean cancer shape index > 1.775) or log standard deviation of the stroma shape index ≤ 0.651 100 / 140 11.57(9.24 – 17.92) Ref 3(Backwards elimination) High risk: PI * > median 86 / 101 8.02(6.08 – 9.86) 1.73(1.26 – 2.37) 0.0007 0.566 Low risk:PI * ≤ median 72 / 102 15.72(9.97 – 25.05) Ref *PI: (0.2701 * log mean stroma area) + (0.3905 * log edge density).