Citations

Abstract Background: Advances in oncology, particularly the adoption of immunotherapies and antibody-drug conjugates, are transforming the nature and frequency of adverse events (AEs). These AEs come with both short and long-term symptoms that significantly impact patient quality of life. Monitoring for early onset of symptoms could optimize personalized therapy, maximizing potential efficacy while mitigating toxicity. It is also possible that some toxicities are directly associated with drug sensitivity. We sought to identify symptoms associated with pathologic complete response (pCR) using patient-reported outcomes (PROs) in early-stage high-risk breast cancer patients. Methods: Our study population included 288 stage II/III high-risk breast cancer patients enrolled on the I-SPY2 trial from 2021-2024, who received IO and ADC neoadjuvant therapies ± standard paclitaxel. pCR was defined as tumor absent in breast and nodes at surgery following neoadjuvant treatment. Patients (n = 288) were sent electronic PROs. 33 patient-reported AEs were measured using NCI's Patient Reported Outcomes - Common Terminology Criteria for Adverse Events (PRO-CTCAE). Each symptom was evaluated using severity, frequency, and interference on a Likert Scale. Presence of early PRO symptoms (cycles 1-3 of treatment) were categorized (at least one of moderate or greater), and odds ratios were computed with pCR as outcome. To assess whether higher grade AEs were enriched in patients that achieved a pCR, we also performed the Wilcoxan Rank Sum test using maximum (worst) symptom severity. Breast tumor volume was measured by MRI at baseline and again at cycle 3 (∼day 21 from treatment start). Functional tumor volume (FTV) was computed from MR images by using enhancement thresholds. and percent change from baseline (ΔFTV) was calculated. Reduction in FTV was defined as a positive change. The Benjamini-Hochberg method was used to adjust for multiple hypothesis testing. Results: Of 288 patients included in our analysis (median age = 48 years, range = 20-78, pCR rate = 29%), 203 (70.5%) were White, 17 (5.9%) were Asian, 33 (11.5%) were Black or African American, and 35 (12.2%) were Hispanic and 89% were administered immunotherapy, 11% ADC. The pCR rate was 29%. PRO analysis revealed that patients that had moderate to severe muscle pain (27% vs 10% OR = 3.15, P < 0.05), joint pain (22% vs 8% OR = 3.23, P < 0.05), headache (27% vs 12.5% OR = 2.59, P < 0.05), or mouth/throat sores (16% vs 5% OR = 3.56, p < 0.05) within weeks 1-3 had higher odds of achieving a pCR. Maximum severity between weeks 1-3 was associated with pCR. Patients that achieved a PCR had higher grade muscle pain (P < 0.05), heart rates of palpitations (P < 0.05), and significantly lower grade numbness and tingling (p = 0.002). Additionally, of the symptoms assessed, frequency, interference, and severity of arm or leg swelling was significantly correlated with reduction in tumor volume (FTV) during weeks 1-3 treatment (Bonferroni corrected P<0.01). Beyond 6 weeks, associations were weak or insignificant. Conclusion: Our study utilizes a computational framework that defined sentinel symptoms such as muscle and joint pain, mouth/throat sores, headache, and swelling, as early as weeks 1-3 that were associated with a favorable tumor response. This may suggest an early immune reaction in patients that eventually respond favorably to treatment, and prioritization to continue these patients on treatment may be warranted. Our work can help provide earlier proactive monitoring to mitigate toxicities, to enhance treatment response, and a potential symptom-based early understanding to personalize treatment efficacy. Citation Format: A. Basu, S. Umashankar, C. Yau, K. J. Ruddy, A. H. Blaes, D. M. Wolf, A. DeMichele, J. Perlmutter, D. Yee, A. B. Olshen, L. J. van't Veer, R. A. Shatsky, C. Isaacs, R. Nanda, H. S. Rugo, M. E. Melisko, N. Hylton, L. J. Esserman, D. L. Hershman, I-SPY Trial Investigators and Patient Advocates.. Early Adverse Symptoms Predict Response to Treatment Among Patients in the I-SPY Trial [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 RF1-04.

Abstract Intro: The 70-gene MammaPrint (MP) assay is prognostic for distant recurrence and predictive of benefit from adjuvant chemotherapy (CT) benefit established on surgical tissue of systemically untreated patients (pts). The MP index range is from -1 to +1 with higher index indicating lower risk of early recurrence. Risk is categorized as MP High Risk, MP Low Risk, and MP UltraLow Risk, and results can guide treatment decisions. Pts with MP index >0 are classified as Low Risk and derive little benefit from CT. There are clinical scenarios in which the only available tissue for MP or other genomic testing has been exposed to neoadjuvant endocrine therapy (NET). Few studies have investigated the impact of ET exposure on MP index. Methods: Pts were identified through the I-SPY Low Risk Registry (LRR), an observational sub-study which enrolled pts with hormone receptor positive (HR+)/HER2-negative early breast cancer ineligible for the I-SPY2 trial due to MP Low Risk or UltraLow Risk status, between 2010-2020. Pts were treated at the discretion of their provider. We conducted a retrospective analysis of MP index from paired tumor tissue obtained from systemically untreated diagnostic core biopsies and surgical specimens after NET. MP categories and index before and after NET were compared. Change in MP index was assessed for an association with changes in clinicopathologic factors obtained through the LRR using the paired-t test. Results: 28 pts with paired tumor samples were included in the analysis. Median age was 48 (range 31-75) and 18 (64%) were premenopausal. At baseline, 13 (46%) were clinically node-positive and 22 (76%) had ductal histology. Tumor grade was 1 in 6 pts (21%), 2 in 20 pts (71%), and 3 in 2 pts (7%). ER expression was >95% in 25 pts (range 60-100%) and PR expression was >90% in 14 pts (range 0-100%). NET regimens included tamoxifen +/- ovarian function suppression (OFS) in 3 pts (10.7%), aromatase inhibitor (AI) +/- OFS in 15 pts (53.6%), tamoxifen followed by or in combination with AI +/- OFS in 8 pts (28.6%), fulvestrant in 1 pt (3.6%) and OFS alone in 1 pt (3.6%). 11 (61%) of premenopausal pts received OFS. 96% (27/28) of pts were classified as Low Risk both pre- and post-NET. Only one pt had MP change from Low Risk to High Risk. Three pts had change from Low Risk to UltraLow Risk or vice versa. As a continuous variable, a small average increase in MP index was observed after NET; but, there was a range of responses. 19 pts (68%) had an increase in MP index after NET, while 9 pts (32%) had a decrease. In those with a decrease in MP index, the average ER expression decreased by -13.6%, whereas in those with an increase in MP index, ER expression was unchanged (-0.18%) (p=0.049). There was no significant difference in changes in PR expression between tumors with increased vs decreased MP indices. At a median follow-up of 35 months, 4 distant recurrences were observed. Of the 9 pts with a decrease in MP index, 3 pts (33%), including the one who changed from Low Risk to High Risk, experienced a distant recurrence. In contrast, in those with an increase in MP index, 1 of 19 (5%) experienced a distant recurrence. Conclusion: In pts with early-stage HR+/HER2- MP Low Risk or UltraLow Risk breast cancer, NET had minimal impact on the tumor MP classification. These data suggest that the MP assay could be used for tissue exposed to NET if untreated tissue is not available. Further investigation in larger cohorts is needed to validate these findings. The observation that there were more distant recurrences in pts with tumors that had an increase in MP after NET is hypothesis generating, and potentially clinically relevant if validated in a larger study. The potential of MP as a dynamic risk marker to stratify recurrence risk following NET is being evaluated in the ongoing I-SPY2 Endocrine Optimization Pilot. Citation Format: S. Jacob, C. Wu, A. Glas, C. Yau, L. Brown-Swigart, G. Hirst, T. Haddad, K. Giridhar, A. Elias, R. Mukhtar, L. Huppert, K. Albain, D. Yee, L. van’t Veer, L. Esserman, J. Chien. Impact of Neoadjuvant Endocrine Therapy on MammaPrint Index in Hormone Receptor-Positive, HER2-Negative MammaPrint Low Risk Early-Stage 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 PS4-05-28.

Abstract Background Invasive lobular carcinoma (ILC) differs from invasive ductal carcinoma (IDC) in its estrogen receptor (ER) signaling pathways, response to systemic therapy, and imaging features. We evaluated differences in early change in the proliferation marker Ki67 and features on magnetic resonance imaging (MRI) after novel endocrine based neoadjuvant therapy in lobular versus non-lobular cases on the I-SPY2 EOP Trial. Methods EOP is a prospective, randomized NET trial enriched for patients with ILC. Patients with stage II-III ER positive HER2 negative breast cancer with low molecular risk (Mammaprint low or Mammaprint Hi1 plus high Sensitivity to Endocrine therapy Index) were randomized to one of several endocrine therapy-based arms. Treatment arms include novel oral SERDS with or without CDK4/6 inhibitor, aromatase inhibitors, and SERMs with or without ovarian function suppression in pre-menopausal patients. Patients undergo repeat core needle biopsy with central Ki67 assessment after 3 weeks of treatment and serial breast MRI. Here, we compare endocrine therapy responsiveness as measured by change in Ki67 at 3 weeks and functional tumor volume (FTV) on pre-surgical MRI in lobular versus non-lobular cases. Results We evaluated 253 patients of whom 148 (58.5%) had non-lobular cancer (93.9% IDC), and 105 (41.5%) had lobular cancer (17.1% mixed ILC/IDC). Overall, mean age was 53.3 years, 63.8% of tumors were grade 2, and 86.5% were Mammaprint (MP) Low with the remaining being Hi1, with higher rates of MP Low in the lobular versus non-lobular cohorts (92.4% versus 83.4%, p=0.0358). At baseline, tumors in the lobular cohort had significantly lower mean Ki67 than those in the non-lobular cohort (11.2% [SD 12.8] versus 16.4% [SD 14.4%], p=0.0038). Consistent with this, the proportion with baseline tumor Ki67 <10% was significantly higher in the lobular cohort than the ductal cohort (60.4% versus 39.3%, p=0.001). At 3-week biopsy, mean Ki67 remained lower in the lobular versus non-lobular group (3.4% versus 5.4%, respectively, p=0.0415). The proportion of patients with Ki67 <10% at 3 weeks did not differ significantly between lobular and non-lobular cases (88.6% versus 79.7%, respectively, p=0.08). Among the 40 lobular cases and 88 non-lobular cases with baseline Ki67 ≥10%, week 3 Ki67 decreased to <10% in 90% of lobular cases and 72.7% of non-lobular cases (p=0.07). On MRI, mean FTV at baseline was similar among lobular and non-lobular cases (21 cc3 versus 16.3 cc3). On pre-surgical MRI, patients in the lobular cohort had significantly higher mean residual FTV than those in the non-lobular cohort (5.7 cc3 versus 3.8 cc3, p=0.034). Interestingly, when splitting the lobular cohort into mixed lobular and pure lobular, mean residual MRI FTV was 7.5 cc3 in mixed cases and 3.8 cc3 in pure lobular cases, reflecting that FTV decreased the least in mixed lobular cases (decrease of 51.1% in mixed cases, 69.7% in pure lobular cases, and 73.9% in non-lobular cases, p=0.0255). Conclusions These results highlight differences in lobular versus non-lobular breast cancer. Lobular cases had lower baseline Ki67, and similar rates of decrease in Ki67 after 3 weeks of NET than non-lobular cases. Despite this, mean FTV after completion of 6 months of NET was higher in lobular tumors. These findings may reflect differences in early versus late indicators of response to NET, or differential response patterns by histology. Alternatively, standard response measures such as Ki67 and FTV may need lobular-specific modifications. Ongoing work includes central assessment of Ki67 from surgical specimens, analysis of changes in gene expression signatures, and evaluation of surgical outcomes such as breast conservation rates by histologic subtype. Citation Format: R. Mukhtar, P. Norwood, A. Borowsky, S. Alkhafaji, K. Giridhar, C. Vaklavas, A. Elias, M. Wei, M. Goetz, N. Onishi, O. Olopade, L. Van 't Veer, L. Huppert, W. F. Symmans, L. Brown Swigart, C. Wu, C. Yau, D. Yee, M. Magbanua, E. Price, N. Hylton, L. Esserman, J. Chien. Differential early response to neoadjuvant endocrine therapy in lobular versus non-lobular breast cancer on the Endocrine Optimization Pilot (EOP) of the I-SPY2 Trial [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 PS4-02-01.

Abstract Background: Invasive lobular carcinoma (ILC) accounts for up to 15% of breast cancer (BC) cases and has distinct clinical and biological features compared to invasive carcinoma of no special type (IBC-NST). Prior studies, including PENELOPE-B, suggested greater benefit from adjuvant palbociclib (palbo) in ILC. We assessed the prognostic and predictive value of histology in patients (pts) treated with endocrine therapy (ET) ± palbo. Methods: PALLAS is a global, phase III trial in which pts with stage II-III, HR+/HER2− BC were randomized to 2 years (y) of adjuvant palbo + ET or ET alone. Pts with IBC-NST or pure ILC per local pathology assessment were included. Endpoints included invasive disease-free (IDFS; primary), distant recurrence-free (DRFS), locoregional recurrence-free (LRRFS), and overall survival (OS). Prognostic value was assessed using univariable and multivariate Cox models adjusted for age, T and N stage, grade, and progesterone receptor. Predictive value was assessed using histology-by-treatment interactions. Results: Among 5,748 pts in the ITT population, 4,982 (87%) were included (3,838 IBC-NST [67%], 1,144 ILC [20%]); median follow-up 6.9y. Compared with IBC-NST, ILC was associated with older age (median 54 vs 51 y), higher T stage (T3/T4: 41.7% vs 20.4%), fewer grade 3 tumors (12.7% vs 33.5%), higher mastectomy rates (74.5% vs 58.4%), less (neo)adjuvant chemotherapy (72.2% vs 86.2%), and less tamoxifen as initial ET (28.8% vs 34.5%) (all p<0.001). ILC prevalence varied by country: lower in Asia (Japan 10.5%, Korea 0%, Taiwan 2.8%) and higher in Europe (23.1%) and North America (24.0%). When evaluating the predictive value, palbo appeared to confer greater benefit in ILC for all endpoints (HRs ranging from 0.41 to 0.86) and not in IBC-NST (HRs ranging from 0.97 to 1.01), although the histology-by-treatment interaction was significant only for LRRFS (Table). In the overall population, 7-year IDFS was 79.3% for IBC-NST and 75.8% for ILC. While the difference was significant in univariable analysis (HR 1.17, 95% CI 1.01-1.35), it was not in multivariable analysis (HR 1.06, 0.91-1.24). Notably, the proportional hazards assumption was violated, and time-dependent modeling demonstrated that worse outcomes for ILC on IDFS, DRFS, and other endpoints emerged predominantly beyond 3 years. Conclusions: In this large, randomized dataset of pts with early HR+/HER2− breast cancer with long-term follow-up, ILC demonstrated distinct clinicopathologic features and geographic distribution. While ILC was not an independent prognostic factor over the entire follow-up, time-dependent modeling revealed worse long-term outcomes which only emerged after 3 years. Palbo appeared to be associated with more favorable outcomes in ILC than in IBC-NST across endpoints. These findings suggest a potential benefit of palbo in ILC that warrants further investigation in other adjuvant trials evaluating CDK4/6 inhibitors. Citation Format: G. Nader-Marta, C. Desmedt, D. Hlauschek, A. DeMichele, G. Zoppoli, E. Blondeaux, E. de Azambuja, M. Bellet-Ezquerra, O. Metzger-Filho, J. M. Suga, G. Pfeiler, M. P. Goetz, M. Ruiz-Borrego, S. Loibl, J. Meisel, A. Ring, K. Van Baelen, E. P. Mamounas, N. Zdenkowski, E. Agostinetto, M. Lambertini, E. Gauthier, L. Soelkner, A. C. Dueck, M. Gnant, E. L. Mayer. Long-term prognostic and predictive value of lobular histology in the PALLAS trial [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 PS3-11-06.

Abstract Background: Combinations of antibody-drug conjugates and immunotherapy, such as trastuzumab deruxtecan (T-DXd) and rilvegostomig (rilve), respectively, offer promising efficacy but are associated with a heightened risk of drug-induced interstitial lung disease (ILD)/pneumonitis. Early detection and management are critical for optimizing patient safety, yet there are no standard protocols for real-time evaluation. We hypothesized that vigilant monitoring would enable early detection, prevent symptoms, and minimize the chance of treatment delays while testing novel therapeutic combinations that have a potential risk of ILD. Methods: We tested the combination of T-DXd and rilve administered every 3 weeks in the I-SPY2.2 phase 2 platform trial (NCT01042379), which enrolls patients with clinical stage II/III breast cancer commencing neoadjuvant systemic therapy. Given the potential risk of overlapping pulmonary toxicity of these two agents, we added rigorous monitoring with pulmonary function tests (PFTs), six-minute walk test (6MWT), and high-resolution chest CT (HRCT) scanning pretreatment and every 6 weeks thereafter. A real-time safety workflow was implemented, with pre-specified criteria for halting treatment and escalation to the next block of treatment. All abnormal findings reported on chest CT by site radiologist were centrally reviewed in real time by the study pulmonologist (T.K.) and cases of suspected ILD were adjudicated promptly by a safety committee consisting of the pulmonologist and several medical oncologists, enabling early study drug interruption and therapeutic interventions. Toxicity data were evaluated in the first 51 patients treated with T-DXd plus rilve, and a preplanned hold was initiated until safety was ascertained by the data and safety monitoring board. Results: Safety data were evaluated in the first 51 patients receiving T-DXd plus rilve. Eight ILD events were confirmed (15.68% incidence): 5 grade 1, 2 grade 2, and 1 grade 3. One patient had an early chest CT scan because of dyspnea. All cases were identified by HRCT scans; PFTs and 6MWT did not enhance early detection and were subsequently dropped from the monitoring protocol. No patient was symptomatic for more than 24 hours. Radiographic findings resolved within 14-55 days. Patients with grade 2 & 3 ILD were treated with corticosteroids. No cases of ILD required intensive care or escalated therapy beyond steroids. Early treatment escalation to the subsequent block of therapy occurred in 3 patients; all 3 patients were subsequently able to receive appropriate therapy. At the meeting, safety data from all 100 patients treated with this combination will be reported. Conclusion: Real-time, centralized monitoring and adjudication allowed for early detection and prompt management of ILD, which both mitigated severity and supported timely therapeutic management. Among the monitoring tools studied, HRCT scanning performed every 6 weeks proved most effective for early ILD identification; PFT and 6MWT did not enhance early detection. This approach offers a potential framework for future studies assessing pulmonary safety in regimens combining agents that have a known risk for inducing ILD. Citation Format: T. Kulkarni, R. Nanda, H. S. Rugo, C. C. OSullivan, K. M. Kalinsky, J. C. Boughey, P. R. Pohlmann, J. Perlmutter, D. Yee, A. Borowsky, F. Symmans, N. Hylton, L. Van ‘t Veer, C. Yau, L. Esserman. Real-time monitoring for drug-induced interstitial lung disease with Trastuzumab Deruxtecan and Rilvegostomig combination therapy prevents persistent, symptomatic disease: Safety insights from the I-SPY2.2 platform trial [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 PS1-01-09.

Abstract Introduction: AC plus a taxane is standard neoadjuvant therapy for high-risk early breast cancer, but AC may be unnecessary for pathologic complete response (pCR) in many patients. I-SPY2 is a phase 2 trial evaluating the use of 12 weeks of paclitaxel +/- an investigational drug followed by AC. Serial measurements of Functional Tumor Volume (FTV) by MRI, inter-regimen biopsies and pathology from surgery were used to determine the contribution of AC to reduction in FTV or pCR. Methods: 1187 patients enrolled between 2010-2022 had serial measurements of FTV. 150 patients received an inter-regimen biopsy. The primary endpoint of I-SPY2 is pCR. To determine how much AC contributed to the reduction in FTV, we calculated the total FTV decrease between pre-treatment and pre-surgery, and then computed the percentage (%) of the total FTV decrease seen after 12 weeks of paclitaxel regimen. We attributed pCRs observed among patients for whom taxane contributed <90% FTV reduction to AC and estimated the % patients achieving pCR from AC. For the subset with inter-regimen biopsy, we estimated the percentage achieving pCR from those patients with invasive cancer present at inter-regimen biopsy. Results: We found that by week 12 (end of taxane), 64% (763/1187) of patients had achieved ≥90% of their total FTV reduction. In 424/1187 (36%) patients had with taxane <90% response, and subsequent AC contributed to more than 10% of the total decrease in FTV. 90 pCRs (21%) were observed among these 424 patients or, overall, approximately 8% of patients achieved pCR from AC after taxane. Of note, 54/1187 (4.5%) patients had no response to a taxane regimen and had decreased in FTV due to AC alone. Of the 54/1187 responding to AC only, there were 11 pCRs. These results varied by response predictive subtypes as shown. When the subpopulation with post-paclitaxel biopsy was examined, 63/150 (42%) patients had invasive cancer at this timepoint. 47 patients received AC and 4/47 (8.5%) achieved a pCR after AC. 13/47 (28%) achieved RCB 0/1. Conclusions: Serial measurements of FTV identifies patients who had a favorable response to a taxane regimen. For those with a suboptimal response, AC provided some contribution to further decrease in tumor volume. However, only a small fraction achieved a complete response. Immune positive and HER2+ (non-luminal) had the biggest benefit, with only 3-6% of other subtypes achieved pCR. Thus, AC can be reserved for those patients with suboptimal response to the initial taxane therapy. I-SPY2.2 utilizes these data to avoid AC by allowing patients to proceed to surgery after serial MRI and biopsy predict a pCR at an early timepoint. While AC remains the standard of care in many neoadjuvant regimens, the low response rates seen in I-SPY2 suggests that more active regimens need to be developed for patients with disease resistant to a taxane based therapy. Citation Format: D. Yee, C. Yau, P. Norwood, D. M. Wolf, P. R. Pohlmann, N. Onishi, L. van 't Veer, A. Borowsky, W. F. Symmans, J. Perlmutter, J. C. Boughey, E. Price, W. Li, I-SPY Investigators, N. Hylton, L. J. Esserman. Quantifying the Contribution of Doxorubicin and Cyclophosphamide (AC) to Pathologic Response Following Neoadjuvant Taxane Therapy: Analysis of the I-SPY2 TRIAL [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 PD6-09.

Abstract Introduction: With emerging antibody drug conjugates (ADC) as treatment of metastatic breast cancer (mBC), it is key to understand the mechanisms that drive resistance to inform optimal ADC selection and sequencing. The AURORA (AUR) US Network was established to characterize paired primary and metastatic (met) samples via multiplatform profiling, with a prospective phase of met tissue collection that included pre and post-ADC samples. Methods: Fresh frozen or FFPE met and archival primary tumors were analyzed across 3 platforms: DNA low pass whole genome and exome sequencing (WES), DNA methylation arrays (Methyl) and RNA sequencing (RNA seq). The molecular cohort included pts with data from ≥1 platform. Pairwise gene expression comparisons were performed in R using t-tests, and pathway analysis with GSEA. Results: A total of 168 pts comprised the molecular cohort: WES, 168; Methyl, 163; RNA seq, 168. Median age at mBC diagnosis was 56 (28-88). 21 pts (12.5%) had known germline BRCA1/2 mutations; 37 (22.0%) presented with de novo mBC. Subtype at mBC (if not available, at primary) diagnosis was: 108 (64.3%) HR+/HER2-, 22 (13.1%) HER2+, 33 (19.6%) TNBC and 5 (3.0%) unknown. Among 139 pts with known subtype in the first AUR met collected, 95 (68.3%) were HR+/HER2-, 17 (12.2%) HER2+ and 27 (19.4%) TNBC. 72 pts received ≥1 ADC with 1 (0-4) median line of chemotherapy for mBC prior to ADC1. As ADC1, 42 pts received HER2 ADC (35 trastuzumab deruxtecan, T-DXd; 7 trastuzumab emtansine), 29 TROP2 ADC (28 sacituzumab govitecan, SG; 1 datopotamab deruxtecan) and 1 NECTIN4 ADC. 22 pts received ≥2 ADC (16 immediate sequence; 6 intervening therapy); 11 T-DXd and 11 SG as ADC2. Median duration of ADC1 was 141 days (6-522); 81 (7-425) for ADC2. Progressive disease was the most common reason for ADC discontinuation, denoting resistance in post-ADC samples. 99 samples were collected pre-ADC1 and 24 post-ADC1, of which 5 were pre-ADC2. ERBB2 expression did not differ between samples after HER2 (n=9) vs non-HER2 ADC (n=8) (p=0.55), nor did TACSTD2 after TROP2 (n=8) vs non-TROP2 ADC (n=9) (p=0.16). In pts with HER2+ subtype at mBC diagnosis or AUR collection who received HER2 ADC, ERBB2 expression significantly decreased in post (n=4) vs pre-treatment (n=7) samples (p=0.043). ERBB2 downregulation was observed in 5 pts with paired pre vs post-HER2 ADC samples. In pts who received TROP2 ADC, TACSTD2 expression did not differ in post (n=10) vs pre-treatment (n=47) samples (p=0.63). GSEA of post-ADC samples revealed significant pathway differences by payload, including actin cytoskeleton in post-microtubule inhibitor (MTi) vs non-MTi ADC samples (p=0.001) and ATM-dependent DNA damage response after topoisomerase I inhibitor (TOP1i) vs non-TOP1i ADCs (p=0.043). Lysosomal vesicle biogenesis (p=0.014) and calcium release (p=0.05) pathways significantly differed between ADC naïve vs treated metastases. ERBB2 hypermethylation was not detected in the cohort. TACST2 promoter methylation was observed in 10 samples (β value >0.3) without association with TROP2 ADC exposure. ERBB2 or TACSTD2 mutations were not identified in the ADC-treated cohort. A novel loss-of-function (LOF) mutation in MED12 was detected in one pt immediately post-T-DXd. MED12 LOF has been reported to promote PARP inhibitor and platinum resistance in BRCA-deficient tumors and may represent a novel resistance mechanism to TOP1i. Conclusions: Multiplatform molecular characterization of mBC treated with ADC revealed potential mechanisms of resistance related to downregulation of target expression, payload (e.g., microtubule formation, DNA damage repair) and lysosomal processing. Additional analyses integrating these multimodal data are ongoing. Strategies to overcome resistance, e.g. modifying ADC target, are being evaluated in clinical trials. Citation Format: A. C. Garrido-Castro, W. C. Nenad, J. M. Balko, T. Hinoue, G. L. Wheeler, B. J. Kelly, M. Singha, B. M. Felsheim, A. R. Michmerhuizen, U. Chandran, A. V. Lee, L. A. Carey, T. A. King, E. Mardis, P. W. Laird, K. V. Giridhar, K. A. Hoadley, C. M. Perou. Molecular characterization of resistance to antibody drug conjugates in metastatic breast cancer: a prospective analysis from the AURORA US Network [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 PD3-09.

Abstract Background: Patients with metastatic breast cancer (MBC) often have short survival, and clinical disease management is challenging due to tumor heterogeneity, treatment resistance, and unique contributions of organ-specific metastatic microenvironments. Methods: The retrospective and prospective phases of the AURORA US Metastasis Project have profiled samples from 220 MBC patients using RNA-seq, DNA-seq, and DNA methylation arrays. RNA-seq has been performed on 392 specimens from 143 primary tumors and 249 metastases. In total, there are 162 primary-metastasis pairs from 134 patients (13 patients have 2+ metastases profiled). The 249 metastatic samples include 116 liver, 29 lymph node, 17 lung, 14 bone, 10 brain, and 8 skin metastases. Each specimen was assayed to determine its clinical and molecular subtypes. Here we investigate the molecular causes of clinical and PAM50 subtype switching in primary tumor-metastasis pairs using multi-platform data. Results: Of the patients with paired data in the AURORA dataset, 38/134 (28%) had clinical subtype (HR+/HER2+, HR-/HER2+, HR+/HER2-, TNBC) switches between the primary tumor and paired metastasis. The most common clinical subtype switch was loss of hormone receptor expression (16/38 patients) in the metastatic tumor. Many of these observations have been confirmed by concomitant RNA and/or DNA changes. In addition, 59/134 (44%) patients had PAM50 molecular subtype switches. 26/59 (44%) patients had a Luminal A primary tumor which switched either into Luminal B (16/26) or HER2-Enriched (7/26) subtypes, suggesting recurrent progression mechanisms during metastasis. For example, patient ARRP had a Luminal A primary tumor and a HER2-Enriched bone metastasis. Using DNA clonality data, we identified a clonal mutation in the DNA binding domain of FOXA1 (H247L) as well as mutations in CDH1 (frameshift) and NLRC3 (Q650H) that originated in the primary tumor. A distinct, dominant clone arose in the metastasis, harboring oncogenic activating mutations in PIK3CA (H1047R) and ERBB2 (L869R). There was also an increase in RNA expression of ERBB2 and FGFR4 but no change in ERBB2 copy number. This is in alignment with previous publications demonstrating a shift to the HER2-Enriched subtype, often without gain of HER2 amplification. Interestingly, there were 21 patients that had both clinical and PAM50 subtype switches. Patient ANFC had a HR+/HER2- primary tumor (Luminal A) and a TNBC liver metastasis (HER2-Enriched). Notably, the metastasis lost clinical expression of ER and PR (by IHC), and this was also observed through a decrease in ESR1 and PGR mRNA expression. Using DNA clonality data, we identified 5 new clones in the metastasis that were not present in the primary tumor. Finally, we observed that basal-like primary tumors with a PAM50 subtype switch in the paired metastasis had a significantly lower correlation to the basal-like centroid compared to primary tumors with a basal-like paired metastasis (p=0.0053). This suggests that either these tumors had significant intra-tumoral heterogeneity in the primary tumor and the metastasis originated from a non-basal-like portion of the tumor, or the switches from a basal-like primary tumor are a byproduct of the PAM50 bioinformatics algorithm (i.e. the tumor is very close to 2 subtype centroids). Ongoing analyses will precisely determine which subtype switches can be attributed to genetic or epigenetic causes, organ site biases, and/or bioinformatic methodologies. Conclusions: Using multi-platform molecular analyses, we identified cases of clinical and molecular subtype switching in primary-metastasis pairs. These findings begin to elucidate the underlying molecular drivers of tumor progression and subtype switching in patients with MBC, which may have implications for clinical treatment. Citation Format: A. R. Michmerhuizen, B. M. Felsheim, G. L. Wheeler, B. J. Kelly, M. Singha, T. Hinoue, W. C. Nenad, A. C. Garrido-Castro, J. M. Balko, K. Giridhar, U. Chandran, A. V. Lee, L. A. Carey, T. A. King, K. A. Hoadley, P. W. Laird, E. R. Mardis, C. M. Perou, AURORA US Metastasis Network. Molecular causes of subtype switching in primary and metastatic breast tumor pairs from the AURORA US Project [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 PD2-06.

Abstract BACKGROUND: EOP is an I-SPY2 sub-study designed to test novel endocrine-based strategies in patients (pts) predicted to have modest benefit from chemotherapy. Vepdegestrant is a selective oral PROTAC estrogen receptor (ER) degrader that has demonstrated efficacy in mutant ER+ HER2-negative MBC. METHODS: EOP eligibility included pts with Stage 2/3 HR+ HER2-negative, MammaPrint (MP) Low Risk BC. Patients with MP High1 signatures were eligible if clinically node-negative or SET 2,3 High. Pts were randomized to one of 3 oral treatment (tx) arms: vepdegestrant 200 mg/d (V), vepdegestrant 200 mg/d +letrozole (VL), vepdegestrant 200 mg/d + abemaciclib 150 mg bid (VA). Pts received tx for six 28-day cycles followed by surgery. Premenopausal (PreMN) pts received ovarian function suppression (OFS) starting C1D1 in the VL arm, and C2D1 in the V and VA arms. The primary endpoint was feasibility, defined as >75% of pts completing >75% of study therapy. Breast MRI functional tumor volume (FTV) and ctDNA (tumor-informed assay) were assessed at baseline (T0), 3 weeks (T1), 12 weeks (T2), and 6 months (T3). Ki67 and ER/PR by IHC were centrally assessed from tumor biopsies at T0, T1, and the surgical specimen. Results: Between 4/2023 and 1/2025, 121 pts were enrolled: 40 in V, 41 in VL, and 40 in VA. Median age was 54 years, 50% preMN. 60% of pts had ductal, 35% lobular, 5% mixed histology. 55% were cN+, and 59% were not breast conservation surgery (BCS) candidates at baseline. Mean baseline Ki67 was 12.5%, 85.1% had MP Low Risk signature, 79.3% were SET 2,3 High. All 3 tx arms met the primary endpoint with 88%, 88%, and 83% completing > 75% study therapy in V, VL, and VA (thus far), respectively. Most common all grade (AG) TRAEs in V and VL arms include fatigue (53% V, 71% VL), hot flashes (48% V, 71% VL), arthralgias (33% V, 49% VL), all G1 or G2 except for 1 G3 fatigue in VL arm. G2 and G3 neutropenia was seen in 1 pt each in V and VL arms. The most common AG TRAEs in VA include diarrhea (80%), fatigue (65%), nausea (58%), hot flashes (50%), neutropenia (43%). High grade TRAEs in VA include neutropenia (23% G3, 5% G4), ovarian cyst requiring surgery (5% G3), diarrhea (15% G3), nausea (3% G3), transaminitis (3% G3), thromboembolic event (8% G3), acute kidney injury (3% G3). Table 1 summarizes key response endpoints by tx arm. CONCLUSION: Six months of neoadjuvant vepdegestrant, alone or in combination with letrozole or abemaciclib, is feasible and demonstrates robust ER/PR degradation. Encouraging anti-tumor activity through reduction of Ki67 and MRI FTV, ctDNA clearance, and achievement of BCS was seen in pre- and postMN women. The neoadjuvant setting provides a rich platform to identify the dynamic range of response and non-response using a multi-modal approach and can be leveraged to inform future trial design and adjuvant therapy. Citation Format: J. Chien, R. M. Mukhtar, C. Yau, A. D. Elias, A. M. Wallace, N. Chan, C. Omene, N. Chen, J. Tseng, M. S. Trivedi, E. Stringer-Reasor, D. Yee, A. S. Clark, A. Thomas, H. Han, M. Arora, C. Nangia, K. Albain, C. Falkson, C. Isaacs, A. Zimmer, M. Rozenblit, L. van't Veer, L. Brown Swigart, G. Hirst, W. Symmans, A. D. Borowsky, N. Onishi, N. Hylton, S. Alkhafaji, K. V. Giridhar, C. Vaklavas, M. Wei, M. P. Goetz, O. Olopade, L. Huppert, C. Wu, L. J. Esserman. I- SPY2 Endocrine Optimization Pilot (EOP): Neoadjuvant vepdegestrant monotherapy or in combination with letrozole or abemaciclib in molecularly selected patients with stage 2/3 HR+ HER2-negative breast cancer (BC) [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 PD10-02.

Background/Objectives: Antigen presenting cells (APCs) and immune cells have unique properties to drive or suppress immune responses. They are therefore key targets for the expression of vaccine antigens or transgene proteins. To better determine the utility of different molecular therapies to modify these cells, mRNA and DNA-based molecular therapy vectors were compared for their ability to genetically modify immune cells after intradermal injections in mice. DNA-based vectors included naked plasmid DNA, plasmid packaged in lipid nanoparticles (LNPs), and replication-defective adenovirus (Ad) vectors. mRNA delivery was mediated by packaging into LNPs like those used in COVID-19 vaccines. Methods: Each vector was used to deliver Cre recombinase into Cre reporter mice whose cells were activated to express green fluorescent protein (GFP) and firefly luciferase after Cre recombination. The mice were injected intradermally (ID) near the base of their tail at a site that drains into the inguinal lymph node. Luciferase activity was imaged in the living mice 1 or 4 days after vector injection. The animals were then euthanized, and luciferase activity was imaged in the draining inguinal lymph node. Cells were prepared from the intradermal injection site and from the draining lymph node to determine which immune cells were genetically modified by phenotyping CD45, CD3, and CD11b GFP-positive cells by flow cytometry. Given that the skin uniquely contains Langerhans dendritic cells, these CD207+ cells were also phenotyped in skin samples and in the draining lymph node. Results: In both the skin and in the draining lymph node, the rank order of luciferase and GFP activation by the vectors were: (1) Ad; (2) mRNA-LNP; (3) DNA-LNP; and (4) naked DNA. Only mRNA-LNP and Ad vectors mediated obvious luciferase activity in the living animals and in the draining lymph nodes by imaging. Notably, both vectors appeared to leak from the ID injection site and not only modify the draining lymph node but also strongly modify the livers of the mice. Naked DNA and DNA-LNP mediated detectable GFP activation in the skin and draining lymph node in some mice, but this activity was low and did not reach statistical significance when compared to PBS-treated animals. mRNA-LNPs and Ad both mediated significant Cre delivery in CD45+, CD3+, CD11b+, and CD207+ immune cells in the skin and in the lymph node, with adenovirus mediating consistently higher levels of expression in all of the tested cells. Conclusions: These data indicate that mRNA-LNP and Ad vectors mediate stronger modification of skin and lymph node immune cells after intradermal injections. Naked DNA and DNA-LNPs were markedly less potent at this activity than the other vectors. These data are consistent with the higher vaccine potency of mRNA-LNP and Ad vectors and suggest that approaches that increase targeting of immune cell subsets may have utility to increase efficacy while also reducing off-target modification of tissues like the liver.