PURPOSECancer clinical trial enrollment remains critically low at 5%-7% of adult patients despite exponential growth in available trials. Manual patient-trial matching represents a fundamental bottleneck, whereas current artificial intelligence (AI) and machine learning patient-trial matching systems lack data standardization and compatibility across health systems. We developed and validated a semiautomated clinical trial patient matching (CTPM) tool to improve recruitment efficiency and scalability.METHODSWe created a hybrid rules-based and natural language processing (NLP)-based pipeline that automatically screens patients using structured and unstructured electronic health record data standardized to the Observational Medical Outcomes Partnership (OMOP) common data model. CTPM performance was first evaluated on one metastatic colorectal cancer (CRC) trial by comparing CTPM accuracy and efficiency to manual chart review. Following the single-trial validation, we then implemented the system across 29 clinical trials spanning multiple cancer specialties and phases.RESULTSFor the single CRC trial, CTPM achieved 94% retrospective and 88% prospective accuracy, matching gold standard clinical chart review with 100% sensitivity. Implementation reduced chart review workload 10-fold and screening time by 41% (3.1 to 1.8 minutes per chart) for those patients who did undergo review. Since September 2022, the system has screened 98,348 patients across 29 trials, identifying 825 eligible candidates and facilitating 117 patient enrollments with 9%-37% consent rates.CONCLUSIONThis AI and NLP tool demonstrates improved efficiency in clinical trial recruitment by enabling research teams to focus on qualified candidates rather than exhaustive chart reviews. The OMOP-based framework supports scalability across health systems, with potential to address enrollment challenges that limit patient access to clinical trials.

BACKGROUND:Tumor-infiltrating lymphocytes (TILs), assessed by visual examination, are prognostic and predictive in early-stage triple-negative breast cancer. Computational assessment may provide a complementary approach. We evaluated the prognostic value of TILs by visual examination and computational assessment. METHODS:Cisplatin vs Paclitaxel for Triple Negative Breast Cancer (TBCRC030; ClinicalTrials.gov identifier NCT01982448) was a randomized phase 2 trial enrolling patients with BRCA1/2-proficient stage I to III triple-negative breast cancer to receive preoperative cisplatin or paclitaxel. The primary endpoint was pathological response at surgery. The TILs were visually scored on digitized pretreatment biopsies per International TILS Working Group recommendations. Computational assessment used the 4D Path QPOR platform to generate TILs, an immune heterogeneity index, and a combined immune/cell cycle biomarker (CmbI). Predictive performance for residual cancer burden 0/1 was assessed using receiver operating characteristic curves and odds ratios (ORs) with 95% CIs; all statistical tests were 2-sided. RESULTS:Of 139 response-evaluable patients, 121 had matched visual examination and computational assessment data (59 on cisplatin, 62 on paclitaxel). Median visual examination TILs were higher in responders (40.0% vs. 10.0%; P = .002) and predicted response (OR = 1.86, 95% CI = 1.24 to 2.87; area under the curve = 0.69, 95% CI = 0.57 to 0.80). Computational assessment CmbI differed by response group and predicted residual cancer burden 0/1 (OR = 3.20, 95% CI = 1.05 to 11.07; area under the curve = 0.62, 95% CI = 0.51 to 0.73). Computational assessment TILs and immune heterogeneity index were not predictive. Visual examination TILs and computational assessment CmbI predicted response to paclitaxel (OR = 2.91, 95% CI = 1.56 to 6.14; OR = 9.17, 95% CI = 2.01 to 66.39, respectively) but not to cisplatin. CONCLUSION:Visual examination TILs and computational assessment CmbI were each associated with response to neoadjuvant chemotherapy in triple-negative breast cancer in the overall cohort and the paclitaxel arm. Computational assessment CmbI did not outperform visual assessment. Further validation is needed before clinical implementation of computational approaches.

Triple-negative breast cancer (TNBC) patients with brain metastases have a poor prognosis and limited treatment options. Preclinical and clinical evidence suggests that radiotherapy may act synergistically with immune checkpoint inhibitors. We conducted an open-label, single-arm, phase II study of atezolizumab plus stereotactic radiosurgery (SRS) in metastatic TNBC patients with brain metastases. The primary endpoint was progression-free survival (PFS) according to the Response Assessment in Neuro-Oncology Brain Metastases (RANO-BM) bi-compartmental model. Secondary endpoints included extracranial objective response rate, overall survival (OS), and safety and tolerability. A safety run-in analysis for dose-limiting toxicity (DLT) was performed after the first 6 patients were enrolled and completed the assessment period. Six patients were enrolled into the safety run-in phase between May 11, 2018 and October 24, 2019. No DLTs were observed, but the study was closed early due to slow accrual. Patients received a median of 2 atezolizumab cycles (range: 2—16), and SRS was administered to all 6 patients. Treatment-related adverse events (TRAEs) occurred in 4 participants (66.7 https://www.clinicaltrials.gov NCT03483012. Trial Open to Accrual: 05/01/2018.

BACKGROUND:Dual anti-human epidermal growth factor receptor 2 (HER2) therapy plus chemotherapy followed by maintenance treatment with HER2-targeted and endocrine therapies is standard first-line treatment for hormone-receptor-positive, HER2-positive metastatic breast cancer. On the basis of preclinical and clinical data, the addition of palbociclib (a selective inhibitor of cyclin-dependent kinases 4 and 6) may overcome resistance to both endocrine and HER2-directed therapies. METHODS:In this phase 3, open-label, randomized trial, we enrolled patients with hormone-receptor-positive, HER2-positive metastatic breast cancer who did not have disease progression after four to eight cycles of chemotherapy plus HER2-targeted therapy. Patients were randomly assigned in a 1:1 ratio to receive maintenance HER2-targeted and endocrine therapies with or without palbociclib. The primary end point was investigator-assessed progression-free survival. Secondary end points included the objective response, clinical benefit, safety, and overall survival. RESULTS:A total of 518 patients underwent randomization: 261 were assigned to receive palbociclib and 257 to receive standard therapy. At a median follow-up of 53.5 months, patients in the palbociclib group had significantly longer progression-free survival than those in the standard-therapy group (median duration, 44.3 months vs. 29.1 months; hazard ratio for disease progression or death, 0.75; 95% confidence interval, 0.59 to 0.96; two-sided P = 0.02). Grade 3 and 4 adverse events, predominantly from neutropenia, occurred in 79.7% and 10.0% of the patients, respectively, in the palbociclib group, as compared with 30.6% and 3.6% of the patients, respectively, in the standard-therapy group. CONCLUSIONS:The addition of palbociclib to maintenance anti-HER2 and endocrine therapies led to a significant improvement in progression-free survival over standard therapy, with increased toxic effects, mainly neutropenia. (Funded by Pfizer and others; PATINA ClinicalTrials.gov number, NCT02947685.).

e13058 Background: Mutations (mut) in the alpha catalytic subunit of PI3K (p110α, PIK3CA gene) occur in ~35-40% of patients with HR+, HER2- metastatic breast cancer (mBC). Until the recent approvals of therapies targeting PI3K/AKT, PIK3CA mut screening was not included in breast cancer clinical biomarker testing guidelines, and cross-assay concordance studies are limited. Here, we compared PIK3CA mut detection among blood and tissue-based PCR and NGS assays used in the first-in-human phase I/Ib trial (NCT03006172) of the recently FDA-approved p110α-inhibitor, inavolisib. Methods: Biomarker eligibility required at least 1 of 17 oncogenic amino acid substitutions in PIK3CA , determined by blood- or tissue-based PCR or NGS testing performed locally at participating sites (184/190 [96.8%] patients) or by Sponsor central testing of tumor tissue with the cobas PIK3CA Mutation Test (cobas PCR) (6/190 [3.2%]). For concordance analyses, available archival or fresh tumor tissue was retrospectively sequenced with the cobas PCR test (n = 111) and/or FoundationOne (n = 155); available pre-study plasma-derived circulating tumor (ct)DNA was sequenced with the FoundationACT or FoundationOne Liquid CDx (n = 131) NGS assays at Foundation Medicine, Inc (FMI ctDNA). Overall concordance – agreement + disagreement for detection of a study-eligible PIK3CA mut between assays – is reported, unless noted otherwise. Results: PIK3CA mut status was confirmed by central cobas PCR for 97/111 (87.4%) local test enrolled patients: 71.4% (10/14) for blood-based tests; 90.2% (46/51) for tissue-based tests; 100% (20/20) for tissue-based PCR; 80.8% (21/26) for other tests. For 14 participants for whom the locally determined PIK3CA mut status was not centrally confirmed, the type of local test included NGS on tissue (5/14), NGS on blood (3/14), PCR on blood (1/14), and unspecified (5/14). PIK3CA mut detection concordance between cobas PCR and FoundationOne tissue testing was 96.2% (100/104). Concordance between cobas PCR and FMI ctDNA testing was 81.1% (77/95). Similarly, the concordance between FoundationOne and FMI ctDNA testing was 80.2% (85/106). Among 18 patients with discordant results between cobas PCR and FMI ctDNA, discrepancies were not attributed to a single assay, with 11 detected by cobas PCR but not FMI ctDNA, and 7 by FMI ctDNA but not cobas PCR. Conclusions: These results, albeit from a small study, demonstrate the ability of both tissue- and blood-based PCR and NGS assays to robustly identify a similar patient population with PIK3CA mut HR+, HER2- mBC, who may benefit from an inavolisib-based therapy regimen. Differences in PIK3CA mut detection concordance between tissue- vs. blood-based assays may reflect mutational profiling of a single-biopsied lesion vs. a multi-lesion representative blood sample, or insufficient ctDNA shedding; reflex testing on an alternative sample may prove useful in such cases.

Background: PARP inhibitors (PARPi) have an approved role in treating patients (pts) with germline BRCA1/2 mutated (gBRCAm) breast cancer in the advanced and early settings, with activity also seen in pts with gPALB2m. Preclinical data suggest PARPi exposure leads to intratumoral activation of the proinflammatory cGAS/STING pathway, recruiting CD8+ T cells, and sensitizing gBRCAm cancers to immune-targeted therapy. Preoperative treatment with PARPi monotherapy has demonstrated notable pathologic complete response (pCR) rates. TBCRC 056, an open-label randomized phase II study, evaluates the PARPi niraparib (N) with the anti-PD-1 antibody dostarlimab (D) in the neoadjuvant treatment of gBRCAm or gPALB2mHER2-negative breast cancer, with cohorts for both triple negative breast cancer (TNBC, Arms A and B) and hormone receptor positive (HR+) disease (Arm C). Results from Arm C are being presented. Methods: Eligible pts in Arm C had a gBRCAm or gPALB2m and a tumor size of >1.0 cm, ER and/or PR+ (>10%). All pts in Arm C received 18 weeks of combination N 200 mg orally once daily and D 500 mg IV every 3 weeks (1 cycle = 3 weeks), with mandatory baseline (BL) and week 3 core tumor biopsies. After 18 weeks, pts either underwent surgery or received additional preoperative systemic therapy (per physician’s choice) if residual disease was present on imaging. The primary study objectives are to evaluate pCR at surgery as well as change in stromal tumor infiltrating lymphocytes (sTILs) from BL to 3 weeks in Arms A and B. Evaluation of pCR and change in sTILs in Arm C are secondary study objectives, with pCR to be reported with a two-sided exact 90% confidence interval without hypothesis testing. Change in sTILs between BL and 3 weeks was evaluated using a Wilcoxon signed-rank test, and the association between sTILs and pCR was assessed using a Wilcoxon rank sum test. Results: Accrual to Arm C was completed after 18 pts with HR+/HER2- breast cancer were enrolled and treated between April 2021 and May 2024 (5 gBRCA1m (27.8%), 13 gBRCA2m (72.2%), 0 gPALB2m). Median age was 41.8 years (range 28.8-61.9), 14 were self-reported White (77.8%), 4 (22.2%) Black, and none Hispanic. Stage distribution was 38.9% stage I, 44.5% stage II, and 16.7% stage III. All 18 pts had ER+ >10% disease. Sixteen of 18 pts have completed protocol therapy through surgery, with results pending for 2 pts. Among evaluable pts, 10 (62.5%) completed 6 cycles of D with mean number of cycles received 4.7. Six pts discontinued D early (1 progressive disease, 1 withdrawal of consent, 4 unacceptable adverse events (AE)). Thirteen pts (81.2%) completed 6 cycles of N with mean number of cycles received 5.4. Three pts discontinued N early (1 progressive disease, 1 withdrawal of consent, 1 unacceptable AE). The most common AE (> grade 2) with D+N were rash (25.0%), elevated liver function tests (18.8%), diarrhea (12.5%), and hypertension (12.5%). Of the 16 pts evaluable for response at this time, 3 had pCR at surgery (18.8%, (90% CI 5.3% - 41.7%)), 11 (68.8%) had residual disease (3 RCB-I, 4 RCB-II, 4 RCB-III), and 2 (12.5%) received additional chemotherapy before surgery. Among 12 pts with evaluable sTILs at BL and 3 weeks, mean sTILs at BL was 11.9% (range 1.0% – 35.0%), mean sTILs at week 3 was 23.8% (range 5.0% - 90.0%), with a mean absolute increase of 11.9%; Wilcoxon signed rank p=0.09). Higher sTILs (as a continuous variable) at BL was associated with pCR (odds ratio 1.09, 95% CI 1.00 – 1.19, Wilcoxon rank sum p = 0.03). Conclusions: In pts with gBRCAm HR+/HER2- breast cancer, 18 weeks of PARPi and anti-PD1 resulted in a pCR rate of 18.8%, and a mean absolute increase of sTILs of 11.9% at 3 weeks. Further data from the TNBC cohort of this study are awaited. Given the reported activity of PD-1 inhibitors in the preoperative management of HR+ breast cancer, additional exploration of targeted non-chemotherapy approaches is of great interest in this pt population. Citation Format: Erica Mayer, Noah Graham, Roberto A. Leon-Ferre, Mariya Rozenblit, Cesar Santa-Maria, Steven Isakoff, Jennifer Specht; Nadine Tung, Vandana Abramson, Jennifer Desrosiers, Beyza Koca, Sara M. Tolaney, Eric P. Winer, Ian E. Krop, Antonio C. Wolff, Geoffrey I. Shapiro, Nabihah Tayob, Jennifer L. Guerriero. TBCRC 056: A Phase II Study of Neoadjuvant Niraparib with Dostarlimab for Patients with BRCA- or PALB2-mutated Breast Cancer: Results From the ER+/HER2- Cohort [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr RF3-01.

Supplementary Figure S1. Study design and tissue availability for biomarker analysis.

Background: T-DXd represents an established treatment option for patients with HER2+ and HER2-low MBC, with relevant activity also shown in HER2-ultralow MBC. No clinical trial data is available to guide treatment after progression on T-DXd. We aimed to produce real-world data to understand the performance of treatment regimens commonly administered after T-DXd. Methods: We conducted a retrospective observational study using the nationwide Flatiron Health electronic health record-derived deidentified database. We included patients with MBC who initiated T-DXd between 12/2019 and 9/2023 and who received an additional line of non-T-DXd anticancer treatment immediately following T-DXd. Tumors were categorized as HER2+ if positive at any timepoint before T-DXd, hormone receptor (HR)+/HER2- or HR-/HER2- (i.e. triple-negative) if never HER2+. Real world progression-free survival (rwPFS) and overall survival (OS) for post-T-DXd treatments were estimated using the Kaplan-Meier method. Results: We identified 633 patients who received a systemic therapy post T-DXd: 352 (56%) with HER2+, 222 (35%) with HR+/HER2- and 59 (9%) with HR-/HER2- MBC. Median age was 58 years, 62.9% of the patients were White, 73.9% were treated in the community setting and 31.0% had de-novo MBC. Median prior lines were 4 in each subgroup (range: 1, 15) and 75% (n=475) had experienced progression while on T-DXd. Outcomes with post-T-DXd treatments were significantly more favorable for patients with HER2+ MBC. Median rwPFS was 4.3 months (mo) for HER2+, 3.0 mo for HR+/HER2- and 2.7 mo for HR-/HER2- MBC (p<0.001), with similar outcomes when restricting to patients with prior progression on T-DXd (rwPFS: 3.9 mo for HER2+, 2.7 mo for HR+/HER2-, 2.7 mo for HR-/HER2- MBC, p<0.001). Median OS was 12.6 mo for HER2+, 8.1 mo for HR+/HER2- and 4.8 mo for HR-/HER2- MBC (p<0.001). Of note, treatment patterns after T-DXd were heterogenous, and outcomes significantly differed by the specific treatment regimen administered (p<0.001). Among patients with HER2+ MBC, the most commonly administered post-T-DXd regimens and median rwPFS were: tucatinib, trastuzumab and capecitabine (n=95, 27.0%) with a rwPFS of 4.7 mo; endocrine treatment (ET)-based regimens (n=57, 16.2%) with a rwPFS of 4.4 mo; chemotherapy + anti-HER2 antibodies (n=56, 15.9%), with a rwPFS of 5.0 mo; anti-HER2 antibodies +/- tyrosine kinase inhibitors (n=34, 9.7%), with a rwPFS of 7.3 mo; T-DM1 (n=27, 7.7%), with a rwPFS of 4.1 mo; and sacituzumab govitecan (SG, n=16, 4.6%), with a rwPFS of 2.3 mo. Among patients with HR+/HER2- MBC, the most commonly administered post-T-DXd regimens and median rwPFS were: SG (n=66, 29.7%) with a rwPFS of 2.5 mo; ET-based regimens (n=46, 20.7%) with a rwPFS of 3.2 mo; taxanes (n=20, 9.0%), with a rwPFS of 3.8 mo; capecitabine (n=18, 8.1%), with a rwPFS of 5.8 mo; and eribulin (n=16, 7.2%), with a rwPFS of 5.9 mo. Among patients with HR-/HER2- MBC, the most commonly administered post-T-DXd regimens and median rwPFS were: SG (n=13, 22.0%) with a rwPFS of 3.0 mo; eribulin (n=9, 15.3%) with a rwPFS of 1.8 mo; multiagent chemotherapy (n=8, 13.6%), with a rwPFS of 2.2 mo; experimental/off label regimens (n=6, 10.2%), with a rwPFS of 1.9 mo; and anthracyclines (n=4, 6.8%), with a rwPFS of 2.8 mo. Across all the 633 patients receiving a post-T-DXd treatment, those that received SG (n=95) experienced shorter rwPFS than those receiving a different regimen (2.7 vs 3.9 mo, p=0.001). Conclusions: In a large real-world database, outcomes of post-T-DXd treatments significantly differed by MBC subtype and type of regimen administered. The use of SG immediately after T-DXd was associated with relatively short rwPFS (≤3 mo) across subtypes, suggesting some degree of cross resistance among topoisomerase 1 ADCs. Citation Format: Paolo Tarantino, Do Lee, Julia Foldi, Pamela R. Soulos, Cary P. Gross, Tess O’Meara, Thomas Grinda, Adrienne G. Waks, Eric P. Winer, Nancy U. Lin, Ian E. Krop, Sara M. Tolaney, Sarah Sammons, Maryam Lustberg. Real-world efficacy of immediate subsequent lines of therapy after trastuzumab deruxtecan (T-DXd) in patients with metastatic breast cancer (MBC) – retrospective study from the nationwide Flatiron database [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr P1-08-08.

Supplementary Figure S4. Decision tree algorithm to construct the multiparameter classifier of response in TBCRC023.

Supplementary Figure S3. Distribution of intrinsic subtypes in (A) TBCRC023 and (B) PAMELA cohort.

BACKGROUND:Early metabolic change on PET/CT was predictive of response to neoadjuvant trastuzumab/pertuzumab (HP) in TBCRC026. We hypothesized that a composite biomarker incorporating PET/CT and HER2 tissue-based biomarkers could improve biomarker performance. METHODS:83 patients with estrogen receptor-negative/HER2-positive breast cancer received neoadjuvant HP alone [pathologic complete response (pCR) 22 %]. PET/CT was performed at baseline and 15 days post initiation of therapy (C1D15). Promising imaging biomarkers included ≥40 % SULmax decline between baseline and C1D15, and C1D15 SULmax ≤3. Baseline tissue-based biomarkers included HER2-enriched intrinsic subtype (72 %, 46/64; NanoString), tumor HER2 protein abundance (median log2 13.5, range log2 7.1-15.9; NanoString DSP), and HER2 3+ (83 %, 64/77; immunohistochemistry). Logistic regressions were fitted to predict pCR with HER2/PET-CT biomarkers. The C statistic assessed overall prediction power. The optimal composite score cut-off was determined by maximizing Youden's index. RESULTS:Factors most predictive for pCR in single predictor models included C1D15 SULmax (OR 0.43; p = 0.007, c = 0.77), % reduction in SULmax (OR 1.03, p = 0.006, c = 0.72) and tumor HER2 protein abundance (OR 1.75; p = 0.01, c = 0.76). The composite of C1D15 SULmax and % reduction in SULmax and their interaction term, had improved probability (c = 0.89 from c = 0.78), with high sensitivity (100 %) and negative predictive value (100 %). The addition of tumor HER2 protein did not further improve prediction power (c = 0.90). CONCLUSION:The HER2/PET-CT biomarker had high prediction power for pCR, however was not superior to the prediction power of PET/CT alone. Non-invasive PET/CT biomarkers may facilitate a response-guided approach to neoadjuvant therapy, allowing intensification and de-intensification of treatment, pending further evaluation.

PURPOSE:Predictive biomarkers to better tailor therapy for patients with early-stage human epidermal growth factor 2 (HER2)-positive breast cancer are a priority. We hypothesized that HER2 and immune-based biomarkers would be predictive of pathologic complete response (pCR) to preoperative trastuzumab/pertuzumab (HP). MATERIALS AND METHODS:Patients with stage II/III, estrogen receptor (ER)-negative, HER2-positive breast cancer received neoadjuvant HP in the TBCRC026 clinical trial. The pCR after receiving HP alone was 22% (18/83). Tumor biopsies were performed at baseline. Secondary correlative objectives were to determine the relationship between HER2-based biomarkers and immune processes with pCR. NanoString code sets BC360 and IO360 were used to compare differential gene expression in baseline tumors that underwent pCR versus no pCR. NanoString GeoMx digital spatial profiling was used to assess immune protein abundance in intraepithelial and stromal segments. Stromal tumor-infiltrating lymphocytes and Ki67 were evaluated by hematoxylin and eosin and immunohistochemistry, respectively. RESULTS:Intraepithelial HER2 protein abundance was significantly associated with pCR (P = .001). Low HER2 abundance tumors were primarily basal-like, and essentially all (19/20) failed to achieve pCR. High HER2 abundance tumors that achieved pCR (14/51) exhibited a high degree of immune cell activity, whereas high HER2 abundance tumors that failed to achieve pCR tumors (37/51) were enriched for M-phase processes and epidermal growth factor receptor signaling. Baseline Ki67 was significantly higher in nonresponders (P = .04). CONCLUSION:ER-negative, HER2-positive breast cancer has unique molecular and immunologic features that may predict pCR after neoadjuvant HP. Validation of these potential biomarkers and composite biomarker analyses may guide design of future clinical trials.

Supplementary Figure S6. Distribution of HER2 gene ratio in relation to PIK3CA mutation status and HER2-enriched subtype.

1004 Background: INAVO, a highly potent and selective PI3Kα inhibitor that also promotes degradation of mutated p110α, is approved by the FDA in combination with PALBO + fulvestrant (FULV) for PIK3CA -mutated, HR+, HER2–, endocrine-resistant advanced BC. HG is a common on-target side effect of PI3K inhibitors. There are limited data for PI3K inhibitors in prediabetic/obese pts. Data from prediabetic/obese pts with HR+, HER2– LA/mBC treated with INAVO from a Phase I/Ib study (GO39374; NCT03006172) are reported here. Methods: Adults ≥ 18 years of age received INAVO alone (Arm A), + letrozole (LET) + PALBO (Arm B), + LET (Arm C), + FULV (Arm D), + FULV + PALBO (Arm E), or + FULV + PALBO + primary prophylactic metformin (Arm F). Data are reported across all arms unless indicated. Pts with baseline risk factors for HG were defined by HbA 1c ≥5.7%, fasting blood glucose ≥ 100 mg/dL, or body mass index ≥ 30 kg/m 2 . Adverse events (AEs) were reported using NCI-CTCAE v4, which utilizes fasting laboratory glucose values for HG severity grading, rather than clinical interventions used in v5. Results: Clinical cut-off was Jan 1, 2024. From190 pts treated, 110 (57.9%) were prediabetic/obese; their median time on INAVO was 222 days (range, 7 to 2,152) and mean cumulative dose intensity was 91.8%. Most prediabetic/obese pts discontinued INAVO due to progressive disease (82 [74.5%]); six (5.5%) discontinued INAVO due to an AE (one due to HG). HG was reported in 80.9% of prediabetic/obese pts (grade 3–4: 34.5%). In pts with two risk factors, 87.9% reported HG (grade 3–4: 39.4%). Among pts with HG, median time to onset was 14 days (range, 1 to 1,674) and 86.0% of events resolved by clinical cut-off. Median time to improvement or resolution of first worst grade ≥ 2 event was 8 days (range, 1 to 64). INAVO dose interruptions, reductions, and discontinuations due to HG were reported in 41.8%, 13.6%, and 0.9% of pts, respectively. The most common anti-HG medications were metformin (52.7%; biguanide; concomitant use in Arm F excluded), empagliflozin (25.5%; SGLT-2 inhibitor), sitagliptin (22.7%; DPP-4 inhibitor), and pioglitazone (13.6%; thiazolidinedione); insulin was used in 8.2% of pts. Median time to metformin start (excluding Arm F) was 14 days (range, 1 to 1,710); the median start dose was 1,000 mg total daily; and the highest daily start dose was 2,000 mg. More than one anti-HG medication was often needed. Conclusions: A high proportion of prediabetic/obese pts were included in GO39374. In most of these pts, HG was manageable with dose interruptions and oral anti-HG medications, most commonly metformin. Data support the use of INAVO in prediabetic/obese pts; further investigation of INAVO in pts with diabetes is warranted. Clinical trial information: NCT03006172 .

Supplementary Figure S5. Decision tree algorithm to construct the multiparameter classifier of response in PAMELA.

Though there has been substantial progress in the development of anti-human epidermal growth factor receptor 2 (HER2) therapies to treat HER2-positive metastatic breast cancer (MBC) within the past two decades, most patients still experience disease progression and cancer-related death. HER2-directed tyrosine kinase inhibitors can be highly effective therapies for patients with HER2-positive MBC; however, an understanding of resistance mechanisms is needed to better inform treatment approaches. We performed whole-exome sequencing on 111 patients with 73 tumor biopsies and 120 cell-free DNA samples to assess mechanisms of resistance. In 11 of 26 patients with acquired resistance, we identified alterations in previously characterized genes, such as PIK3CA and ERBB2, that could explain treatment resistance. Mutations in growing subclones identified potential mechanisms of resistance in 5 of 26 patients and included alterations in ESR1, FGFR2, and FGFR4. Additional studies are needed to assess the functional role and clinical utility of these alterations in driving resistance.

542 Background: Invasive lobular carcinoma (ILC) is the second most common histologic subtype of breast cancer (BC), representing 10–15% of cases. HER2 overexpression is rare in ILC, and there is limited data on the clinical characteristics and outcomes of patients (pts) with HER2-positive (HER2+) ILC treated with adjuvant trastuzumab. This study aims to investigate the prognostic value of ILC histology in this setting. Methods: ALTTO was a multicenter, randomized phase III trial evaluating the efficacy of trastuzumab, lapatinib, their sequence, or combination as adjuvant therapy in pts with HER2+ early BC. Pts with pure ILC or invasive BC of no special type (NST) who were enrolled in trastuzumab-containing arms of the ALTTO trial were included in this analysis. Central pathology review confirmed histologic subtype and was used for classification, while local pathology was used when centralized review was unavailable (USA and China). Survival outcomes, including disease-free survival (DFS), and overall survival (OS) were evaluated using Kaplan-Meier method and multivariate Cox regression adjusted for prognostic factors. Patterns of relapse were analyzed and compared across histological subtypes. Time to distant recurrence (TTDR) and time to CNS recurrence were summarized using cumulative incidence functions. Results: Among pts in the trastuzumab-containing arms (N = 6281), 84.4% underwent central pathology review, with a concordance rate of 67.4% for ILC diagnosis. A total of 61 pts with pure ILC (1.0% of the cohort) and 5981 pts with NST were included in the analysis. Pts with ILC were older (mean 54.8 vs. 50.9 years; p=0.002), more likely White (95.1% vs. 68.8%; p<0.001), and postmenopausal (72.1% vs. 56.3%; p=0.01). The proportion of pts with ILC (vs NST) was higher in Europe (67.2% vs 53.7%) and lower in Asia-Pacific (8.2% vs 30.6%) (p<0.001). A significantly higher proportion of ILC (vs NST) were hormone receptor-positive (80.3% vs. 57.4%; p<0.001), Grade 1-2 (51.7% vs. 39.3%; p=0.05). At a median follow-up of 9.8 years (IQR 6.9-10.0), no significant differences in DFS (hazard ratio [HR] 1.14, 95% CI 0.66-1.97; adjusted HR [aHR] 1.33,0.77–2.31), OS (HR 0.96, 0.43-2.15; aHR 1.09, 0.48–2.44), or TTDR (HR 1.67, 0.91–3.05) were observed between ILC and NST. Central nervous system (CNS) relapses were more frequent in ILC (13.6% at 10y, 95% CI 7.1–26.1%) than in NST (5.0%, 4.5–5.7%), with an HR of 3.14 (1.52–6.48) for CNS recurrences in ILC when compared to NST. Conclusions: Long-term outcomes were comparable between ILC and NST in HER2+ early BC treated with trastuzumab-containing regimens. The higher incidence of CNS metastases in ILC highlights its unique relapse pattern, necessitating further investigation to optimize treatment. High discordance between central and local pathology emphasizes the need for standardized histological review in trials and treatment decisions. Clinical trial information: NCT00490139 .

1065 Background: ESR1 mutations ( ESR1 mut) commonly drive acquired resistance to estrogen deprivation by aromatase inhibitors, a first-line standard of care for HR(+)HER2(-) metastatic breast cancer (MBC). We previously published that approximately 63% of patients with HR(+)HER2(-) MBC at progression have a liquid biopsy (LBx) negative for ESR1 mut. The absence of an ESR1 mut may either accurately reflect the tumor genotype (true negative) or represent a false negative due to insufficient ctDNA shedding, with the risk of missing an actionable mutation. Among these patients, 40% exhibit a high ctDNA TF ≥1% (informative negative), while 60% have a low ctDNA TF < 1% (indeterminate negative). This suggests that up to 38% of all patients with HR(+)HER2(-) MBC in this context could potentially benefit from reflex tissue biopsy (TBx) for ESR1 mut in cases deemed indeterminate negative by LBx due to low ctDNA shedding. The goal of this study is to determine the rate of ESR1 mut detection in a new TBx after an indeterminate negative result from FoundationOne Liquid CDx (F1LCDx). Methods: This study included a cohort of patients with BC who underwent tissue and liquid Foundation Medicine comprehensive genomic profiling (CGP) within an interval of up to 90 days during routine clinical care. Clinical data of a subset of patients with confirmed HR(+)HER2(-) MBC was obtained from the US-wide deidentified Flatiron Health-Foundation Medicine MBC clinicogenomic database (CGDB). The data originated from ~280 cancer clinics (~800 sites of care) between 01/2014-09/2024. False negative rate (FNR) and positive percent agreement (PPA) for ESR1 mut detection were calculated with tissue CGP as reference. Results: A total of 522 BC patients underwent TBx and LBx. Among these, 229 (43.9%) had ctDNA TF < 1%. Without accounting for TF, the overall FNR for ESR1 mut was 6.3% and the PPA was 67.1%. In LBx with TF ≥1%, the FNR for ESR1 mut was 0.9% and PPA was 96.0%. In contrast, for TF < 1% samples, the FNR was 12.0% and the PPA 25.7%. 101 patients were included in the CGDB and had a confirmed HR(+)HER2(-) MBC, in which 56 (55.4%) had LBx with ctDNA TF < 1%. The overall FNR for ESR1 mut in this subset of patients was 9.5% and the PPA was 61.9%. In LBx with TF ≥1%, the FNR was 0% and PPA was 100%. And for TF < 1%, the FNR was 15.1% and PPA 20.0%. Conclusions: BC patients with informative negative ESR1 mut (defined as LBx ESR1 mut negative with TF ≥1%) are unlikely to have ESR1 mut detected on tissue CGP testing. However, patients with indeterminate negative ESR1 mut (defined as LBx ESR1 mut negative with TF < 1%), 12-15% were found to be false negatives. This suggests that approximately 5% of all HR(+)HER2(-) MBC patients with ESR1 mut could be missed without reflex testing with a TBx. ctDNA TF levels offer critical guidance in deciding when a reflex to tissue is warranted, ensuring accurate treatment.

3055 Background: Risk of recurrence for patients (pts) with HR+/HER2- breast cancer persists for decades. Most distant recurrences occur in the ‘late’ adjuvant setting, > 5 years (yrs) from diagnosis. In CHiRP (ASCO 2022), we showed that minimal residual disease (MRD) was detectable in the late adjuvant setting: ctDNA was detected in 8/83 (9.6%) pts in the cohort and 6/8 (75%) pts with positive ctDNA (+ctDNA) had developed distant recurrence when initially reported (median follow-up 2 years from first plasma sample collected on study). Here, we report updated clinical outcomes and investigate the meaning of a ctDNA test result during surveillance with longer follow-up. Methods: In CHiRP, pts with stage II-III HR+/HER2- breast cancer at high risk of recurrence diagnosed > 5 yrs prior with no evidence of recurrence were prospectively identified. All pts provided informed consent for prospective plasma collection every 6-12 months at routine follow-up visits for batched, retrospective ctDNA testing using RaDaR, a tumor-informed whole exome sequencing-based assay. Pts were followed at the discretion of the clinical provider without any routine surveillance imaging, as per guideline-concordant care. See CHiRP ASCO 2022 presentation for additional methods. Results: Of 83 pts in the analytic cohort, 57 (68.7%) pts had stage III disease, and most (n = 75, 90.4%) underwent (neo)adjuvant chemotherapy. All pts received endocrine therapy. In this update, median follow-up from first sample collection was 4.4 yrs (interquartile range 4.0, 4.9). 214 plasma samples were collected prior to any known recurrences and included in this analysis. 8/83 (9.6%) pts had +ctDNA at any timepoint including 4/83 (4.8%) with +ctDNA on first study plasma sample. In pts initially ctDNA-negative (-ctDNA; n = 4), median time from first sample collection to MRD detection was 1.29 yrs (range, 0.72 – 3.05). During follow-up, 8 (9.6%) pts developed distant recurrence and 1 (1.2%) pt had a local recurrence. With additional follow-up included in this update, all 8/8(100%) pts with +ctDNA developed distant recurrence with a median lead time of 1.39 years (range 0.01 – 4.24). Among -ctDNA plasma samples with > 2 yrs of follow-up (n = 185), the negative predictive value (NPV) of a -ctDNA test for lack of clinical recurrence for > 2 yrs post-test was 98.4% (3/185). The NPV indicating freedom from recurrence > 1 and > 3 yrs was 100% (0/196) and 96.6% (5/147), respectively. Conclusions: In pts with high-risk HR+/HER2- breast cancer in the late adjuvant setting, all pts with +ctDNA developed distant metastasis. A -ctDNA test was strongly associated with lack of recurrence over a 3 yr follow-up period. Future studies are needed to determine if ctDNA-guided intervention can impact clinical outcomes for early-stage breast cancer and to determine the optimal role of MRD surveillance during follow-up.

TPS1137 Background: Antibody-drug conjugates (ADCs) have demonstrated substantial improvement in progression free survival (PFS) and overall survival (OS) in phase III clinical trials in patients with metastatic triple negative breast cancer (mTNBC) and hormone receptor positive/HER2-negative (HR+/HER2-) metastatic breast cancer (MBC), offering an effective new treatment strategy. Several outstanding questions impact the use of these drugs clinically, and prospective real-world data is needed. First, it is important to understand the safety and efficacy of these agents in a real-world population with diverse patient characteristics. Second, it is critical to understand the safety and efficacy of these ADCs in sequence. Third, it is essential to identify biomarkers that can help clarify mechanisms of response and resistance to ADCs, which may inform future sequencing and treatment strategies. Methods: This is a multicenter prospective registry study of patients with HER2-negative MBC who are treated with sequential ADCs per standard of care (SOC) with the goal to understand the safety and efficacy of sequential ADCs in a real-world setting (NCT06774027). A total of 100 participants with HER2-negative MBC will be enrolled in this study, either prior to starting their first ADC per SOC (cohort 1 = HR+/HER2-; cohort 2 = mTNBC) or prior to starting their second ADC per SOC (cohort 3 = HR+/HER2-; cohort 4 = mTNBC). The dual primary endpoints are real-world progression free survival (rwPFS) of ADC1 and rwPFS of ADC2. Secondary endpoints include overall response rate (ORR), duration of response (DOR), best overall response (BOR), disease control rate (DCR), and real-world overall survival (rwOS), and safety for each ADC. Exploratory endpoints include translational correlates of response/resistance to ADCs (e.g., circulating tumor DNA, circulating tumor cells, and tissue spatial correlates) and patient-reported outcomes (PROs). rwPFS and rwOS will be estimated by the Kaplan-Meier method. Statistics will be descriptive. Enrollment to start in the first quarter of 2025. Clinical trial information: NCT06774027 .