Hormone therapies are frequently used to reduce breast cancer risk in individuals at increased risk for primary or subsequent disease; however, tissue-level responses to these therapies are heterogeneous and incompletely understood. Background parenchymal enhancement (BPE) on breast magnetic resonance imaging (MRI) provides a non-invasive radiologic readout of breast tissue features associated with endocrine responsiveness and cancer risk. Although BPE is associated with hormonal exposure, a subset of patients with BPE do not show a response to preventive endocrine therapy and therefore may remain at increased breast cancer risk. In this study, we integrated single-nucleus RNA sequencing and spatial transcriptomics to define the determinants of endocrine responsiveness in the setting of BPE. We identify hormone-driven epithelial cells with high levels of estrogen signaling and endocrine responsiveness, together with immune-associated epithelial programs characterized by diminished luminal identity and increased expression of immune-modulatory pathways, including major histocompatibility complex (MHC) class II and CD74. Functional organoid assays validate that these epithelial states exhibit differential sensitivity to tamoxifen and demonstrate that inflammatory signals can induce immune-modulatory epithelial programs. Together, our findings identify hormone signaling and immune programs as key determinants of endocrine responsiveness in breast tissue and provide a biological basis for interpreting radiologic markers relevant to cancer prevention.

Neoadjuvant chemotherapy (NAC) may allow de-escalation of axillary surgery; yet treatment disparities persist. We aimed to assess race-based disparities in use of axillary lymph node surgery (ALND) among patients who achieve a nodal response in the context of a large, multicenter NAC trial. We conducted a retrospective analysis of the I-SPY 2 trial. All patients received NAC, but type of surgery was not mandated. Multivariable logistic regression was used to predict odds ratio (OR) of undergoing ALND by race while adjusting for clinical and demographic confounders, including age, region, tumor receptor subtype, clinical and pathologic node status (cN and ypN +/−, respectively), and clinical and pathologic tumor size (cT and ypT, respectively). Among 1394 patients, 79.4

Background: Patient-derived organoids (PDOs) generated from benign breast tissue and breast carcinomas have successfully recapitulated their in vivo counterparts. PDOs model tumorigenesis and allow for screening novel therapeutics personalized to individual patients. However, acquiring cells to generate PDOs is cumbersome. This study demonstrates the feasibility of fine-needle aspiration biopsy (FNAB) for harvesting cells for PDOs modeling ductal carcinoma in situ (DCIS) and compares the efficacy with core needle biopsy (CNB). Methods: Surgical specimens from patients with biopsy-proven DCIS were used for this study. CNB was performed on fresh specimens in the operating room, and tissue was mechanically dissociated before culture in basement membrane extract (BME) and organoid medium to generate PDOs. FNAB was performed in the pathology gross room on fresh specimens, and the aspirate was similarly submitted for culture. Results: PDOs were successfully generated in 15 of 18 specimens obtained by CNB and seven of 11 specimens obtained by FNAB. The average time to initial organoid growth was 4 days for FNAB specimens compared to 19.3 days for CNB specimens. Tumor cells were seen on seven of 11 FNAB smears and 16 of 18 CNB touch preps. Immunofluorescence staining confirmed the presence of both luminal and myoepithelial cells in derived PDOs. Conclusions: FNAB effectively obtains cells for PDOs modeling DCIS. CNB yielded PDOs with a high success rate, but they were slow to establish. The time to organoid growth was significantly shorter for FNAB specimens. Thus, FNAB offers an efficient alternative for breast PDO culture and can reduce the time and resources spent on generating PDO cultures.

Supplementary Table 1. ILC patient and tumor characteristics, including cases with missing data.

Image-based features of the normal breast such as mammographic density and background parenchymal enhancement (BPE) on contrast-enhanced breast magnetic resonance imaging (MRI) are independent breast cancer risk factors. Both are associated with estrogen signaling and respond to endocrine therapy. Yet, molecular and cellular mechanisms underlying these risk factors remain unclear. We aimed to investigate such mechanisms with the goal of identifying novel approaches for cancer risk stratification and personalized prevention. Single-nucleus RNA sequencing on a 10x Chromium platform was performed on normal human breast tissues from 28 women undergoing prophylactic mastectomies (n = 11), lumpectomies (n = 10), or mastectomies for breast cancer (n = 7) who consented to our biobanking protocol. Differences in cell composition and signaling pathways between individuals with different breast densities and BPE levels were determined. Additionally we assessed estrogen responsiveness in patient-derived organoid (PDOs) from women with different levels of BPE using qRT-PCR. Evaluation of immune infiltration as potential mediator of BPE is underway using immunofluorescence and spatial transcriptomics. 12, 377 nuclei from 28 women were analyzed, including 16 with low (minimal or mild) and 12 with high (moderate or marked) BPE levels on MRI, and 5 cases with not dense and 23 with dense breasts on mammography. All main breast epithelial and stromal cell types were represented in our data. We observed great heterogeneity in cell composition between patients. Samples from BPE-high patients contained luminal secretory cells with reduced expression of differentiation markers and luminal hormone sensing cells with higher expression of estrogen response genes like PGR and GREB1. In agreement with this, we observed increased response to estrogen signaling in PDOs from high (n = 5) versus low BPE cases (n = 5). Moreover, luminal cells from BPE-high samples expressed significantly higher levels of CD74 which has been correlated with immune infiltration in cancer. Supporting this, interactions between luminal cells with macrophages via CD74 and APP/COPA were predicted in BPE-high but not in low samples. Comparing samples with distinct mammographic densities, we observed a higher proportion of luminal secretory cells, basal cells, and fibroblasts in patients with high mammographic densities suggesting that distinct mechanisms mediate BPE levels and breast density. Our findings indicate that BPE is influenced by two distinct mechanisms: estrogen-mediated changes in cellular composition and chronic inflammation, both potentially increasing breast cancer susceptibility. Understanding the basis of modifiable breast cancer risk factors will enable risk stratification and personalized treatment, ultimately reducing breast cancer incidence in high-risk populations. Nadine Goldhammer, Marin Bont, Michael Choi, Shruti Warhadpande, Heather Greenwood, Soumi Gottipati, Tam Binh Bui, Michael Bruck, Shoko Emily Abe, Michael Alvarado, Cheryl Ewing, Karen Goodwin, Rita Mukhtar, Jasmine Wong, Laura van't Veer, Nola Hylton, Laura J. Esserman, Jennifer M. Rosenbluth. Changes in breast hormone signaling and inflammation underlie radiologic features of cancer risk [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2367.

While the impact of socioeconomic factors on breast cancer diagnosis, treatment, and outcomes are well-documented, few studies have focused on invasive lobular carcinoma (ILC), the second most common type of breast cancer. We evaluated the relationships between race and socioeconomic status (SES) with clinicopathological characteristics and outcomes in patients with stage I-III ILC using the National Cancer Database (NCDB). We used the NCDB, a national oncology database, to evaluate insurance status, a composite measure of SES (education and income), clinicopathological characteristics, and outcomes in patients with stage I-III ILC. Clinicopathologic variables included tumor size, presence of lymphovascular invasion (LVI), and tumor receptor subtype (hormone receptor, HR), and tumor grade. Overall survival was analyzed with multivariable Cox proportional hazards models. We identified 269,657 patients with stage I-III ILC. Patients in the Medicaid/no insurance group and those with lower SES had larger tumors, more positive lymph nodes, fewer HR+ tumors, and higher-grade tumors. Those in the low SES group had higher rates of chemotherapy use and, in those with HR+ tumors, lower rates of endocrine therapy use. In a multivariable model adjusting for SES, self-identified race/ethnicity, age, stage, receptor subtype, grade, treatment, and Charlson–Deyo score, patients with low SES had a 24

Background: The risks of postoperative complications in breast cancer patients vary by patient and tumor characteristics. Elevated BMI and invasive lobular carcinoma (ILC) increase risks of surgical complications and positive margins, respectively. Methods: We retrospectively analyzed patients with BMI >30 kg/m(2) from an institutional ILC database. The primary outcome was surgical complication rate by procedure type. The secondary outcome was positive margin rates by surgical approach, stratified by T stage. Results: Of 154 analyzed patients, standard BCS, lumpectomy with oncoplastic closure, and simple mastectomy had the lowest complication rates (18.2 %, 17.0 %, 11.8 %). Oncoplastic reduction mammoplasty and mastectomy with aesthetic closure had the highest rates (35.5 %, 33.3 %). The overall positive margin rate was 28.5 %, significantly higher in BCS vs. mastectomy (37.4 % vs. 15.0 %, p = 0.003). Oncoplastic surgery significantly reduced positive margin rates in BCS. Conclusion: In this study, 23.4 % of patients experienced surgical complications, with higher rates in oncoplastic/reconstructive approaches. However, oncoplastic surgery reduced positive margins, highlighting the importance of balancing risks for optimal surgical planning.

Background Patient-derived organoids (PDOs) generated from benign breast tissue and breast carcinomas have successfully recapitulated their respective in vivo counterparts. PDOs model tumorigenesis and allow for screening of novel therapeutics personalized to individual patients. However, acquiring cells to generate PDOs is cumbersome. We demonstrate the feasibility of fine needle aspiration biopsy (FNAB) for harvesting cells for PDOs modeling ductal carcinoma in situ (DCIS). Methods Surgical specimens from patients with biopsy-proven DCIS were used for this study. Core needle biopsy (CNB) was performed on fresh specimens in the operating room, and tissue was mechanically dissociated before culture in basement membrane extract (BME) and organoid medium to generate PDOs. FNAB was performed in the gross room on fresh specimens, and the remaining aspirate was similarly submitted for PDO culture. Results PDOs were successfully generated in 15/18 specimens obtained by CNB and 7/11 specimens obtained by FNAB. The average time to initial organoid growth was 4 days for FNAB specimens compared to 19.3 days for CNB specimens. Tumor cells were seen on 7/11 FNAB smears and 16/18 CNB touch preps. Immunofluorescence staining confirmed the presence of both luminal and myoepithelial cells in derived PDOs. Conclusions FNAB effectively obtains cells for PDOs modeling DCIS. CNB after mincing yielded PDOs with a high success rate, but they were slow to establish. Notably, the time to organoid growth was significantly shorter for FNAB specimens. Thus, FNAB offers an efficient alternative for breast PDO culture and can reduce the time and resources spent on generating PDO cultures. ### Competing Interest Statement The authors have declared no competing interest. National Institutes of Health, https://ror.org/01cwqze88, R01CA281361 UCSF Resource Allocation Program

Abstract Approximately 80% of ductal carcinoma in situ (DCIS) and invasive breast cancers express hormone receptors (HR), and are typically treated with selective estrogen receptor modulators (SERMs) or aromatase inhibitors, or with additional HR-targeting agents in the research setting. However, the development of resistance poses a significant challenge. To better understand the mechanisms behind endocrine resistance and explore new treatment strategies, it is essential to have a reliable model system that accurately reflects the heterogeneous nature of breast lesions. In this study, we developed a protocol to generate patient-derived organoid cultures of pre-invasive and malignant breast lesions from patients with persistent or residual disease in the active surveillance or neoadjuvant setting after treatment with endocrine therapy. We generated n = 11 HR+ DCIS and n = 18 HR+ breast cancer organoids and used single-cell RNA-sequencing to characterize cell heterogeneity within individual organoid cultures. We observed mixed non-cancerous and malignant cells present within some cultures, and optimized protocols to minimize non-cancerous cell types. To better define endocrine sensitivity versus resistance, we also assessed responsiveness to estrogen and anti-estrogen therapy in individual cultures by monitoring organoid growth through live cell imaging and evaluating estrogen-response markers at the protein and RNA levels in response to treatment in vitro. Next, we applied our model system to evaluate endocrine resistance in heterogeneous tumor populations. We compared changes in cellular composition and signaling pathways in endocrine-resistant and endocrine-sensitive breast organoids based on in vitro metrics and/or correlation with clinical variables. In addition, within single cultures heterogeneous cell populations were identified with differential expression of ESR1 and HR target genes, generally falling into three cluster groups that could be defined based on different patterns of FOXA1, GATA3, TFF1, and XBP1 expression, among others. These populations also differed with evidence of co-expressed genes and signaling pathways related to breast tumorigenesis and differentiation (estrogen response, androgen response, mTOR signaling, metabolism) in putative HR-active cell populations in both pre-invasive and invasive breast cases. In summary, our study demonstrates that patient-derived breast and breast cancer organoids can be used to model estrogen responsiveness and endocrine resistance. We have established a platform that enables us to investigate the mechanisms underlying endocrine resistance at different premalignant and tumor stages to evaluate new treatment approaches. The ultimate goal of this work is to contribute to the development of new targeted therapies for endocrine-resistant breast lesions. Citation Format: Nadine Goldhammer, Shruti Warhadpande, Liana Beld, Cristian K. Maldonado Rodas, Allison Lam, Tam Binh Bui, Shoko Emily Abe, Michael Alvarado, Cheryl Ewing, Karen Goodwin, Rita Mukhtar, Jasmine Wong, Lamorna Brown Swigart, Gillian Hirst, Christopher Schwartz, Laura van't Veer, Laura Esserman, Jo Chien, Jennifer M. Rosenbluth. Heterogeneous cell populations in organoid models derived from breast preinvasive and invasive carcinomas after endocrine therapy [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Breast Cancer Research; 2023 Oct 19-22; San Diego, California. Philadelphia (PA): AACR; Cancer Res 2024;84(3 Suppl_1):Abstract nr A059.

Purpose Establishing breast MRI imaging patterns associated with neoadjuvant immunotherapy is needed to monitor response. We analyzed serial breast MRIs in patients receiving neoadjuvant chemo-immunotherapy on the ISPY-2 clinical trial. Methods Patients had stage 2–3 HER2-negative breast cancer in a single institution. Regimens included: weekly paclitaxel (control), weekly paclitaxel in combination with pembrolizumab, or weekly paclitaxel in combination with pembrolizumab and intra-tumoral injection of SD-101, a TLR9 agonist. All patients then received AC and underwent surgery. Regional lymph nodes were retrospectively evaluated on breast MRI at baseline, 3 wks, 12 wks and 20 wks by a single blinded radiologist. MRIs were assessed for development of new regional lymphadenopathy, or increase in longest diameter or cortical thickness of largest regional lymph node. Results Between 12/2015-4/2021, a total of 43 patients enrolled to the control (n = 16) and paclitaxel + pembrolizumab +/- SD-101 (n = 27) arms. 25 patients had hormone-receptor positive disease and 18 patients had triple negative disease. 12 of 27 patients (44.4%) receiving chemo-immunotherapy experienced increased lymphadenopathy within the first 12 weeks compared to 1 of 16 patients (6.3%) in the control group (p = 0.014). Increased lymphadenopathy was observed despite concomitant decrease in breast tumor size at all time points. 11 of 12 patients with increased lymphadenopathy had pathologically negative nodes at surgery. There was no association between lymphadenopathy and lower residual cancer burden (p = 0.696) or immune-related toxicity. Conclusions Chemo-immunotherapy was associated with early increases in regional lymphadenopathy despite decreased breast tumor size. Increased lymphadenopathy was not associated with node-positive disease at surgery.

Among the goals of patient-centric care are the advancement of effective personalized treatment, while minimizing toxicity. The phase 2 I-SPY2.2 trial uses a neoadjuvant sequential therapy approach in breast cancer to further these goals, testing promising new agents while optimizing individual outcomes. Here we tested datopotamab-deruxtecan (Dato-DXd) in the I-SPY2.2 trial for patients with high-risk stage 2/3 breast cancer. I-SPY2.2 uses a sequential multiple assignment randomization trial design that includes three sequential blocks of biologically targeted neoadjuvant treatment: the experimental agent(s) (block A), a taxane-based regimen tailored to the tumor subtype (block B) and doxorubicin-cyclophosphamide (block C). Patients are randomized into arms consisting of different investigational block A treatments. Algorithms based on magnetic resonance imaging and core biopsy guide treatment redirection after each block, including the option of early surgical resection in patients predicted to have a high likelihood of pathological complete response, the primary endpoint. There are two primary efficacy analyses: after block A and across all blocks for the six prespecified breast cancer subtypes (defined by clinical hormone receptor/human epidermal growth factor receptor 2 (HER2) status and/or the response-predictive subtypes). We report results of 103 patients treated with Dato-DXd. While Dato-DXd did not meet the prespecified threshold for success (graduation) after block A in any subtype, the treatment strategy across all blocks graduated in the hormone receptor-negative HER2-Immune-DNA repair deficiency- subtype with an estimated pathological complete response rate of 41%. No new toxicities were observed, with stomatitis and ocular events occurring at low grades. Dato-DXd was particularly active in the hormone receptor-negative/HER2-Immune-DNA repair deficiency- signature, warranting further investigation, and was safe in other subtypes in patients who followed the treatment strategy. ClinicalTrials.gov registration: NCT01042379. In the I-SPY2.2 trial, patients with high-risk stage 2/3 breast cancer received neoadjuvant datopotamab-deruxtecan, followed by sequential chemotherapy with or without targeted therapy, with the option of early surgical resection after each block of therapy. In a subgroup of patients, the sequential treatment strategy was superior to standard of care.

Background Invasive lobular carcinoma (ILC) of the breast is known for high risk of late recurrence, yet some patients still recur within 5 years of diagnosis. Determining factors associated with early/late recurrence could help tailor treatment and surveillance strategies. Methods Using an institutional database, we evaluated patients with ILC and ≥ 5 years of follow-up or recurrence within 5 years. We used multivariate logistic regression and the Kaplan-Meier method to evaluate which clinicopathologic features and treatment strategies were associated with recurrence < 5 years since diagnosis versus recurrence ≥ 5 years since diagnosis. Additionally, we explored the association between Clinical Treatment Score 5 (CTS5) with early versus late recurrence. Results Among 513 cases of stage I–III ILC, there were 75 early and 54 late recurrences during a median follow-up period of 9.4 years. Early recurrence was associated with larger tumors (mean 4.2 cm vs. 2.9 cm, p < 0.0001), higher incidence of > 3 positive nodes (32.4% vs. 9.11%, p > 0.0001), and more aggressive tumor biology (low/negative progesterone receptor expression, higher grade, and higher Ki67). Late recurrence was associated with younger age (mean 55.6 vs. 59.2 years, p = 0.037) and elevated body mass index (BMI > 25 kg/m 2 in 60.1.0% vs. 45.4%, p = 0.021). Omission of adjuvant endocrine therapy or radiotherapy after lumpectomy conferred increased risk of early rather than late recurrence. Conclusion Factors related to tumor aggressiveness and treatment were associated with early recurrence, whereas patient related factors were related to late recurrence. These data may help guide treatment strategies and surveillance approaches for patients with ILC.

Background: Treatment of estrogen receptor (ER)-positive breast cancer with selective estrogen receptor degraders (SERDs) frequently results in the loss or reduction of ER expression. Whether these changes are due to on-target effects of SERDs degrading ER or arise as a mechanism of tumor resistance with associated changes in cellular phenotypes remains unknown. It is critical to distinguish between these possibilities to accurately assess treatment response and determine the most appropriate subsequent therapy. To this end, we created and conducted molecular analyses on patient-derived organoid cultures from post-treatment tissue in patients receiving neoadjuvant SERD therapy for early-stage ER+ breast cancer in the I-SPY2 Endocrine Optimization Protocol (EOP). Methods: The I-SPY2 EOP study is a prospective, randomized substudy within the I-SPY TRIAL testing the oral SERD amcenestrant alone or in combination with letrozole or abemaciclib in stage 2/3 ER+ Her2-negative breast cancer. Enrollment is ongoing, with patients receiving amcenestrant neoadjuvantly for 6 months until the day before surgery. Tumor tissue is collected at baseline, 3 weeks, and at surgery. Organoids were generated from post-treatment surgical samples. Organoid cultures were optimized based on established methods (Dekkers et al., Nature Protocols, 2021) to assess ER levels and activity. Pre- and post-treatment tissue samples were also assessed for ER, PR, Ki67, and GATA3, a luminal marker and transcription factor that is functionally linked with ER, via immunohistochemistry. Results: In 7 patients with both pre- and post-treatment tissue samples including fresh surgical samples for organoid generation, the ER in baseline tumor tissue was >=90% in all patients, PR ranged from 40-90%, and Ki67 ranged from 5-30%. In post-treatment surgical tissue from these cases, ER ranged from 0-30%, PR from 0-50%, Ki67 from < 1%-10%, and GATA3 was positive in 5 of 5 cases tested to-date. The creation of organoids from residual disease at surgery was attempted for these 7 patients, with organoids successfully propagated in 5 cases thus far. 3 of 5 organoid cultures were ready for analysis and in all cases strong ER and PR expression in organoids was observed after culture for > 1 month in the absence of amcenestrant. Detailed gene expression profiling (including Mammaprint/Blueprint) and gene set enrichment analyses (GSEA) to assess for intrinsic breast cancer subtype and ER activity in each sample and corresponding organoid culture are in progress and will be reported with the full dataset. Conclusion: Patient-derived organoid culturing of residual disease after neoadjuvant endocrine therapy is feasible. Neoadjuvant treatment with a SERD can render ER and PR low or absent at the time of surgical resection, which does not necessarily imply the presence of endocrine therapy resistant disease. The use of organoids and additional IHC markers (GATA3) demonstrate that receptor negativity may be an indicator of the drug hitting its target, suggesting ER signaling is still intact. In general, patient-derived tumor organoid cultures modeling residual disease states can be a useful adjunct to existing methods used to monitor the effects of neoadjuvant endocrine therapy and is being explored in the I-SPY EOP trial. Citation Format: Jennifer Rosenbluth, Christopher J. Schwartz, Tam Binh Bui, Shruti Warhadpande, Pravin Phadatare, Sigal Eini, Michael Bruck, Julissa Molina-Vega, Kami Pullakhandam, Nicole Schindler, Lamorna A. Brown Swigart, Christina Yau, Gillian Hirst, Rita Mukhtar, Karthik V. Giridhar, Olufunmilayo I. Olopade, Kevin Kalinsky, Cheryl A. Ewing, Jasmine M. Wong, Michael D. Alvarado, Laura Van’t Veer, Laura J. Esserman, Jo Chien. Characterization of residual disease after neoadjuvant selective estrogen receptor degrader (SERD) therapy using tumor organoids in the I-SPY Endocrine Optimization Protocol (EOP) [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P3-09-01.

Supplementary Figures and Tables except Table S7

BackgroundAxillary surgery after neoadjuvant chemotherapy (NAC) is becoming less extensive. We evaluated the evolution of axillary surgery after NAC on the multi-institutional I-SPY2 prospective trial.MethodsWe examined annual rates of sentinel lymph node (SLN) surgery with resection of clipped node, if present), axillary lymph node dissection (ALND), and SLN and ALND in patients enrolled in I-SPY2 from January 1, 2011 to December 31, 2021 by clinical N status at diagnosis and pathologic N status at surgery. Cochran-Armitage trend tests were calculated to evaluate patterns over time.ResultsOf 1578 patients, 973 patients (61.7%) had SLN-only, 136 (8.6%) had SLN and ALND, and 469 (29.7%) had ALND-only. In the cN0 group, ALND-only decreased from 20% in 2011 to 6.25% in 2021 (p = 0.0078) and SLN-only increased from 70.0% to 87.5% (p = 0.0020). This was even more striking in patients with clinically node-positive (cN+) disease at diagnosis, where ALND-only decreased from 70.7% to 29.4% (p < 0.0001) and SLN-only significantly increased from 14.6% to 56.5% (p < 0.0001). This change was significant across subtypes (HR-/HER2-, HR+/HER2-, and HER2+). Among pathologically node-positive (pN+) patients after NAC (n = 525) ALND-only decreased from 69.0% to 39.2% (p < 0.0001) and SLN-only increased from 6.9% to 39.2% (p < 0.0001).ConclusionsUse of ALND after NAC has significantly decreased over the past decade. This is most pronounced in cN+ disease at diagnosis with an increase in the use of SLN surgery after NAC. Additionally, in pN+ disease after NAC, there has been a decrease in use of completion ALND, a practice pattern change that precedes results from clinical trials.

AbstractBackground: Although investigators have shown associations between socioeconomic status (SES) and outcomes in breast cancer, there is a paucity of such data for invasive lobular carcinoma (ILC), the second most common type of breast cancer. Herein we evaluated the relationship between SES with tumor features and outcomes in stage I to III patients with ILC. Methods: We analyzed a prospectively maintained institutional ILC database and utilized the area deprivation index (ADI) to determine neighborhood adversity, an indicator of SES. We used Cox proportional hazards models in Stata 17.0 to evaluate relationships between ADI quintile (Q), race, body mass index (BMI), clinicopathologic features, treatment type, and event-free survival (EFS). Results: Of 804 patients with ILC, 21.4% lived in neighborhoods classified as ADI Q1 (least resource-deprived) and 19.7% in Q5 (most resource-deprived). Higher deprivation was significantly associated with larger tumor size (3.6 cm in Q5 vs. 3.1 cm in Q1), increased presence of lymphovascular invasion (8.9% in Q5 vs. 6.7% in Q1), and decreased use of adjuvant endocrine therapy (67.1% in Q5 vs. 73.6% in Q1). On multivariable analysis, tumor size, receptor subtypes, and omission of adjuvant endocrine therapy were associated with reduced EFS. Conclusions: These data show that patients with ILC and higher ADI experience more aggressive tumors and differences in treatment. More data evaluating the complex relationships between these factors is needed to optimize outcomes for patients with ILC, regardless of SES. Impact: ADI is associated with differences in patients with ILC.

Neoadjuvant chemotherapy (NAC) increases rates of successful breast-conserving surgery (BCS) in patients with breast cancer. However, some studies suggest that BCS after NAC may confer an increased risk of locoregional recurrence (LRR). We assessed LRR rates and locoregional recurrence-free survival (LRFS) in patients enrolled on I-SPY2 (NCT01042379), a prospective NAC trial for patients with clinical stage II to III, molecularly high-risk breast cancer. Cox proportional hazards models were used to evaluate associations between surgical procedure (BCS vs mastectomy) and LRFS adjusted for age, tumor receptor subtype, clinical T category, clinical nodal status, and residual cancer burden (RCB). In 1462 patients, surgical procedure was not associated with LRR or LRFS on either univariate or multivariate analysis. The unadjusted incidence of LRR was 5.4% after BCS and 7.0% after mastectomy, at a median follow-up time of 3.5 years. The strongest predictor of LRR was RCB class, with each increasing RCB class having a significantly higher hazard ratio for LRR compared with RCB 0 on multivariate analysis. Triple-negative receptor subtype was also associated with an increased risk of LRR (hazard ratio: 2.91, 95% CI: 1.8-4.6, P < 0.0001), regardless of the type of operation. In this large multi-institutional prospective trial of patients completing NAC, we found no increased risk of LRR or differences in LRFS after BCS compared with mastectomy. Tumor receptor subtype and extent of residual disease after NAC were significantly associated with recurrence. These data demonstrate that BCS can be an excellent surgical option after NAC for appropriately selected patients.