Highlands Oncology Group

Abstract Background: Brain metastases (BrM) are a frequent and clinically challenging manifestation of HER2-positive (HER2+) metastatic breast cancer (mBC), affecting up to 50% of patients (pts) over the course of their disease. Current guidelines recommend trastuzumab deruxtecan (T-DXd) as the preferred second-line therapy for HER2+ mBC overall, while the combination of tucatinib, trastuzumab, and capecitabine (TTC) remains an important option for pts with active or progressing BrM disease due to its proven intracranial activity. However, there is a lack of real-world data comparing treatment outcomes with these regimens specifically among pts with BrM involvement. This study evaluates real-world time to next treatment (TTNT) in pts with prior HER2+ therapy and documented BrM treated with second-line T-DXd vs TTC. Methods: We used the IntegraConnect PrecisionQ de-identified electronic health record database, which includes data on over 3 million cancer pts across more than 500 care sites, to identify individuals with HER2+ mBC who received either T-DXd or TTC as second-line therapy on or after January 1, 2020. Eligible pts had documented BrM prior to initiating second-line treatment. HER2+ status was inferred based on receipt of at least one HER2-targeted therapy during first-line metastatic treatment. Individuals who received T-DXd in their first-line of metastatic therapy were excluded from the analysis. The index date was defined as the start of second-line therapy with either TTC or T-DXd. Descriptive statistics were used to summarize demographic and clinical characteristics by treatment group, including age at index date, race, ethnicity, health insurance, BMI, ECOG performance status, estrogen receptor status, first-line metastatic regimen (grouped as ado-trastuzumab emtansine, trastuzumab (H) +pertuzumab (P) + taxane, endocrine therapy +/- HP, or other regimens), and presence of liver, bone, or lung metastases. TTNT was analyzed using Kaplan-Meier survival curves, with median estimates reported. Multivariable Cox proportional hazards regression was used to evaluate the association between treatment and TTNT, adjusting for the aforementioned demographic and clinical covariates. Hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) are presented. Results: A total of 255 pts met study eligibility, of whom 138 (54%) received T-DXd as second-line therapy. Pts treated with T-DXd were slightly older than those who received TTC (median age: 58 vs 55 years; Wilcoxon p=0.026). A lower proportion of pts in the T-DXd group were classified as overweight (BMI 25-30) compared to the TTC group (17% vs 29%; chi-square p=0.026). No other significant differences in demographic or clinical characteristics were observed between treatment groups. Median TTNT was significantly longer among pts who received T-DXd compared to TTC (17 vs 11 months; log-rank p=0.006). This difference remained significant after adjustment for demographic and clinical covariates, with T-DXd associated with a 48% reduced likelihood of requiring subsequent therapy (HR: 0.52; 95% CI: 0.37-0.74; p<0.001). Conclusions: In this real-world analysis of pts with prior HER2+ therapy and documented BrM, second-line treatment with T-DXd was associated with significantly longer TTNT compared to TTC among pts with BrM. These findings may help inform treatment decisions in this high-risk population. However, interpretation is limited by the retrospective design, potential selection bias, and the use of treatment patterns to infer HER2 status. Most importantly, granular clinical data on BrM, including prior local therapies (radiation and/or surgery) and CNS-specific outcomes, were unavailable. Further prospective studies are needed to validate these findings. Citation Format: V. Gorantla, R. Choksi, F. Kudrik, D. Patt, S. Reddy, S. Reganti, S. Rosenfeld, G. Cioffi, T. Sura, D. Parris, A. Rui, M. Gart, C. Wall, B. Wang, P. Varughese, J. Donegan, L. Morere, R. Geller, J. Scott, R. Mahtani. Real-world time to next treatment in HER2+ breast cancer patients with brain metastases: trastuzumab deruxtecan vs tucatinib based therapy as second-line treatment [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 PS5-01-12.

Abstract Background: According to recent findings from the DESTINY-06 trial, HR-positive (HR+)/HER2-low metastatic breast cancer (mBC) patients (pts) who received trastuzumab deruxtecan (T-Dxd) following ≥1 line of endocrine therapy (ET) experienced improved progression-free survival (PFS) compared to those who received chemotherapy (chemo) in a second-line (2L) setting. To assess whether similar benefits are seen in a real-world setting, we evaluated overall survival (OS) and time to next treatment (TTNT). TTNT was used as a surrogate measure for PFS. Methods: We used the IntegraConnect PrecisionQ de-identified electronic health record database of >3 million cancer pts across >500 care sites, to identify pts. Eligible pts for the study were adults with a confirmed diagnosis of HR+/HER2-low mBC who had ≥2 documented visits and a CDK4/6 inhibitor(i) +/- ET in the first-line metastatic setting. Pts with a subsequent treatment (2L) with T-Dxd or chemo were identified and observed between 6-1-2022 and 4-30-2025. TTNT was defined as time from the 2L start to either the initiation of a subsequent line of therapy or death. We did not assess for the reason pts discontinued (e.g., toxicity or disease progression). TTNT and OS were analyzed using Kaplan-Meier analysis. Hazard ratios (HR) adjusting for age at treatment and race were estimated using Cox proportional hazard models. Results: Among 340 eligible pts, 82 received TDxd and 248 received chemo in the 2L setting. The median age at treatment was 64 (interquartile range 55, 70) years old with a composition of 70.0% White, 10.6% African American. Among pts with a reported Eastern Cooperative Oncology Group (ECOG) status at treatment, 91.5% had an ECOG score of ≤1. Median OS was not significantly different for pts who received T-Dxd compared to chemo (Table), but median TTNT was significantly longer for pts who received T-Dxd (10.1 months, 95% CI: 7.8, 13.1) compared to chemo (6.0 months, 95% CI: 5.1, 6.5). Pts using T-Dxd were significantly less likely to change treatment compared to chemo after adjusting for age, and race [HR (95% CI): 0.45 (0.30, 0.66), P<0.01)]. Treatment crossover occurred in both cohorts: 28% of pts in the chemo group utilized T-Dxd in a later line of therapy, and 19% of pts in the T-Dxd group utilized chemo in a later line of therapy. Conclusions: Consistent with findings from a subset of the Destiny-06 trial, HR+/HER2-low mBC pts previously treated with a CDK4/6i who then utilized T-Dxd in the 2L setting were associated with a significantly improved TTNT compared to 2L chemo utilization. OS was not statistically different between T-Dxd and Chemo. These results may be impacted by unmeasured factors such as, tumor burden, endocrine resistance, and access to care. Further studies are needed to confirm the optimal sequencing of TDxd in the real-world setting. Citation Format: R. Mahtani, R. Choksi, F. Kudrik, D. Patt, S. Reddy, S. Reganti, S. Rosenfeld, S. W. Champaloux, D. Parris, A. Rui, A. Sharma, M. Gart, C. Wall, B. Wang, P. Varughese, J. Donegan, L. Morere, R. Geller, J. Scott, V. Gorantla. Real-world comparison of time to next treatment for patients with HR-positive, HER2-low metastatic breast cancer treated with either chemotherapy or trastuzumab deruxtecan in the second-line [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 PS1-11-17.

Adavosertib is a highly selective, small molecule Wee1 inhibitor that sensitizes tumor cells to cytotoxic agents. We report results from the lead-in cohorts of two phase II studies: carboplatin/pemetrexed plus adavosertib versus carboplatin/pemetrexed in first-line metastatic non-squamous non-small-cell lung cancer (NSCLC) (NCT02087241); and docetaxel plus adavosertib versus docetaxel in recurrent NSCLC (NCT02087176). Both lead-in cohorts assessed early safety and efficacy (objective response rate [ORR]). First-line metastatic treatment was carboplatin (area under the curve to 6 h [AUC6]) plus pemetrexed 500mg/m2 intravenously on day 1, every 21 days; recurrent treatment was docetaxel 75mg/m2 intravenously on day 1 plus prophylactic granulocyte-colony stimulating factor 6mg subcutaneously on day 4 every 21 days. All patients received adavosertib 225mg twice daily orally on days 1-3 (five doses). After a planned safety analysis of the first-line trial, dose and schedule were modified to reduce toxicity. First-line: 14 patients were enrolled in four treatment cohorts. Median time on trial was 17.3 weeks, with an ORR of 29%. The most common adverse events were diarrhea (50%), nausea (50%), vomiting (50%), anemia (43%), neutropenia (43%), decreased appetite (43%), and dehydration (43%). Recurrent: 32 patients were enrolled. The ORR was 9%. The most common adverse events were diarrhea (66%), anemia (50%), nausea (47%), fatigue (47%), vomiting (44%),and thrombocytopenia (41%). Both studies terminated early; the recurrent study after an interim analysis showed increased toxicity and limited efficacy, and the first-line study after a change in first-line standard of care. ClinicalTrials.gov, NCT02087241 and NCT02087176.

Việc tìm ra các dẫn chất mới kháng ung thư đang là vấn đề đáng được quan tâm hiện nay. Các dẫn chất chứa nhóm chức ure và các dẫn chất 5-methoxy-2,2-dimethyl-2H-chromen đã cho thấy có tác dụng kháng ung thư tốt trên các dòng tế bào ung thư khác nhau. Với mục tiêu tìm kiếm các hoạt chất mới có khả năng kháng tế bào ung thư, trong nghiên cứu này một số dẫn chất (5-methoxy-2,2-dimethyl-2H-chromen-6-yl)-urea (VIa, VIb, VIc) đã được tổng hợp. Các chất mục tiêu này được tổng hợp từ phản ứng giữa hợp chất trung gian phenyl (5-methoxy-2,2-dimethyl-2H-chromen-6-yl) carbamat và các amin tương ứng trong môi trường kiềm (TEA), sử dụng ACN làm dung môi. Dẫn chất trung gian phenyl (5-methoxy-2,2-dimethyl-2H-chromen-6-yl) carbamat được tổng hợp từ 5-fluoro-2-nitrophenol thông qua năm bước phản ứng cơ bản có hiệu suất cao, gồm phản ứng thế nucleophil, phản ứng ngưng tụ đóng vòng, phản ứng alkyl hóa, phản ứng khử và phản ứng tạo carbamat. Các chất tổng hợp được khẳng định cấu trúc bằng các phương pháp phổ khác nhau, bao gồm phổ khối, phổ hồng ngoại, phổ cộng hưởng từ hạt nhân 1H và 13C-NMR. Các chất mục tiêu được thử hoạt tính gây độc tế bào trên hai dòng tế bào ung thư vú MDA-MB-231 và ung thư phổi A549. Kết qủa cho thấy cả ba dẫn chất đều thể hiện tác dụng gây độc tế bào trung bình trên dòng tế bào MDA-MB-231, trong khi các dẫn chất VIb, VIc không thể hiện hoạt tính trên dòng tế bào A549. Hợp VIa thể hiện tác dụng tốt nhất ở cả hai dòng tế bào thử nghiệm (IC50 = 22,89 µM với dòng tế bào MDA-MB-231 và 16,08 µM với dòng tế bào A549, tuy nhiên tác dụng yếu hơn so với chứng dương doxorubicin. Điều này cho thấy khi gắn nhóm thế vào nguyên tử nitro còn lại của nhóm chức ure trong khung cấu trúc (5-methoxy-2,2-dimethyl-2H-chromen-6-yl)-urea, nhóm thế quinolin-8-yl cho tác dụng kháng tế bào ung thư tốt hơn các hợp chất gắn nhóm thế 4-florophenyl hoặc 2.3-dihydro-1H-inden-2-yl.

TPS4618 Background: Urothelial carcinoma (UC) has a high mortality rate in patients (pts) with metastatic disease. While immune checkpoint inhibitors (CPI), including programmed cell death protein 1 (PD-1) inhibitors combined with chemotherapy (CTx) or enfortumab vedotin (EV), have been approved for first-line treatment of metastatic (m)UC, many pts have de novo or develop acquired resistance. For pts without response to frontline treatment or whose disease has progressed on prior CPI combinations, optimal treatment is unclear and new therapies are urgently needed. Glycoprotein A repetitions predominant (GARP) is a membrane-bound receptor that complexes with latent transforming growth factor (TGF)-β1; the release of active TGF-β1 from this complex suppresses antitumor responses. Livmoniplimab (livmo), an antibody targeting the GARP:TGF-β1 complex, prevents release of active TGF-β1, thereby promoting antitumor activity. A first-in-human phase 1 study (NCT03821935) demonstrated that combining livmo and the anti–PD-1 mAb budigalimab (budi) resulted in a manageable safety profile and promising antitumor activity in pts with PD-1–refractory advanced UC (J Clin Oncol 2024;42[suppl 4]: abs 617). Herein, we describe the phase 2 study that is evaluating livmo + budi vs CTx in pts with mUC (NCT06632951). Methods: This multicenter, open-label, randomized study is enrolling pts aged ≥18 years who have mUC, measurable disease per RECIST v1.1, ECOG PS 0–1, and have experienced disease progression on anti–PD-1 or anti–PD-1 ligand 1 therapy. Platinum (Pt)-eligible pts must have received a Pt-containing regimen; pts who can receive EV must have experienced disease progression on/after receiving EV treatment. Primary objectives are to identify the recommended phase 3 livmo dose in combination with budi and evaluate overall survival. Secondary objectives include evaluating progression-free survival, best overall response of complete or partial response, duration of response, and assessment of safety and tolerability, pharmacokinetics, and immunogenicity of the combination. Pts will be randomized 1:1:1 to 3 arms: 1) livmo dose 1 Q3W + budi Q3W; 2) livmo dose 2 Q3W + budi Q3W; or 3) investigator’s choice of CTx (paclitaxel, docetaxel, or gemcitabine). Pts will be stratified by ECOG PS (0 vs 1) and first-line therapy (pembrolizumab + EV vs CTx). Treatment for pts in Arms 1 and 2 will continue until a maximum of 35 cycles. For pts in Arm 3, treatment will continue for the duration that is consistent with local guidelines/practice for this pt population. No crossover between arms will be permitted. For all pts, treatment is discontinued at disease progression or if other protocol-defined discontinuation criteria are met. In total, approximately 150 pts (50 pts/arm) are planned for enrollment globally. Clinical trial information: NCT06632951 .

LBA13 Background: Neoantigen-targeting immunotherapy aims to provide therapeutic benefit to patients by generating potent neoantigen-specific T cells. GRANITE is an immunotherapy regimen that uses chimpanzee adenovirus and self-amplifying mRNA vaccines encoding patient-specific neoantigens in combination with immune checkpoint inhibitors. Preliminary efficacy was noted in MSS-CRC (Palmer et al Nature Medicine 2022) and this regimen is being assessed in a randomized Phase 2 study in 1L metastatic MSS-CRC (NCT05141721). Methods: Patients (pts) with 20 neoantigens based on the EDGEÔ prediction model were randomized 1:1 to receive the GRANITE regimen in addition to 1L standard of care (SOC) (GRANITE vaccine arm), or 1L SOC alone (control arm). Results: Baseline demographics and disease characteristics were balanced between arms. As of 15 Oct 2024, GRANITE pts had an improvement in PFS relative to control pts (HR=0.73, 90% CI [0.44, 1.21]). In this study, high and low disease burden groups were identified using baseline ctDNA assessed by Gritstone’s validated, highly sensitive, tumor-informed assay. Patients with low disease burden benefited more than those with high disease burden, consistent with emerging data suggesting neoantigen-targeting immunotherapy having greater activity in minimal disease settings (e.g., adjuvant setting). Furthermore, in the low disease burden group at the most recent study visit, more patients were free of progression with very low ctDNA levels (ie, below the limit of quantification of the assay) in the GRANITE arm, 41% (7/17), versus the control arm, 21% (3/14). The greater proportion of GRANITE patients both free of progression and with very low ctDNA, suggests potential further separation of PFS cuves with additional follow up. Neoantigen-specific T cell responses were observed in 100% of patients in the GRANITE arm with evaluable PBMC samples (17/17 assessed by ex vivo and in vitro stimulation ELISpots). Molecular responses based on a reduction in ctDNA at a single timepoint was similar between arms (39% in GRANITE and 40% in control arms). Most common (≥20%) treatment-related adverse events (TRAE) included pyrexia and influenza like illness and no pts discontinued due to a TRAE. Conclusion: This randomized study shows GRANITE may benefit patients, especially those with low burden disease, in this front line setting of metastatic MSS-CRC that has not benefitted from immunotherapy. Updated analysis will be presented. Clinical trial information: NCT05141721 .

International audience

Purpose:Seizure-related homolog protein 6 (SEZ6) is a novel target expressed in small cell lung cancer (SCLC). ABBV-011, a SEZ6-targeted antibody conjugated to calicheamicin, was evaluated in a phase I study (NCT03639194) in patients with relapsed/refractory SCLC. We report initial outcomes of ABBV-011 monotherapy.Patients and Methods:ABBV-011 was administered intravenously once every 3 weeks during dose escalation (0.3–2 mg/kg) and expansion. Patients with SEZ6-positive tumors (≥25% of tumor cells with ≥1+ staining intensity by IHC) were preselected for expansion. Safety, tolerability, antitumor activity, and pharmacokinetics were evaluated.Results:As of August 2022, 99 patients received ABBV-011 monotherapy [dose escalation, n = 36; Japanese dose evaluation, n = 3; dose expansion, n = 60 (1 mg/kg, n = 40)]; the median age was 63 years (range, 41–79 years). Also, 32%, 41%, and 26% of patients received 1, 2, and ≥3 prior therapies, respectively. The maximum tolerated dose was not reached through 2.0 mg/kg. The most common treatment-emergent adverse events were fatigue (50%), nausea (42%), and thrombocytopenia (41%). The most common hepatic treatment-emergent adverse events were increased aspartate aminotransferase (22%), increased γ-glutamyltransferase (21%), and hyperbilirubinemia (17%); two patients experienced veno-occlusive liver disease. The objective response rate was 19% (19/98). In the 1-mg/kg dose-expansion cohort (n = 40), the objective response rate was 25%; the median response duration was 4.2 months (95% confidence interval, 2.6–6.7); and the median progression-free survival was 3.5 months (95% confidence interval, 1.5–4.2).Conclusions:ABBV-011 1.0 mg/kg every 3 weeks monotherapy was well tolerated and demonstrated encouraging antitumor activity in heavily pretreated patients with relapsed/refractory SCLC. SEZ6 is a promising novel SCLC target and warrants further investigation.

2603 Background: Myeloid cells contribute to suppression of adaptive immunity within the TME and limit the efficacy of immune checkpoint inhibitors (ICIs) in head and neck squamous cell carcinoma (HNSCC). Semaphorin 4D (SEMA4D) signaling through its receptors (PlexinB1/B2, CD72) promotes recruitment and suppressive function of myeloid suppressor cells (MDSC). In preclinical and clinical studies, SEMA4D antibody blockade attenuated MDSC and increased penetration and organization of dendritic cells (DC) and T cells into tertiary lymphoid structures that enhanced activity of ICI. We hypothesize that SEMA4D blocking antibody pepinemab may regulate infiltration and crosstalk of immune cells in TME as a novel and complementary mechanism of immune enhancement when combined with immune checkpoint therapy. Methods: KEYNOTE-B84 (NCT04815720) is an ongoing single-arm open-label study evaluating the safety, efficacy, and PK/PD of pepinemab in combination with pembrolizumab as first-line treatment of recurrent or metastatic HNSCC. Exploratory biomarker analyses were performed to evaluate spatial interactions of tumoral immune cells. Pre- and on-treatment tumor biopsies were collected and assessed by multiplex immunohistochemistry for up to 36 biomarkers/biopsy. Unbiased algorithms identified co-localization of markers for advanced cell phenotyping, density, spatial and proximity analysis. Biomarker results were then stratified by demographic and clinical outcome measures. Results: An increase in activated APC (HLA-DR+CD11c+ and HLA-DR+CD68+) and reduced density of MDSC (Arg1+CD14+ and ARG1+CD15+) was observed in patients with durable disease control. Interestingly, spatial analysis of tumor biopsies revealed that combination therapy induced the formation of highly organized immune aggregates, including a high density of activated B cells, DCs, CD4+ and CD8+ T cells, including stem-like CD8+TCF1+PD1+ T cells. Presence of immune aggregates increased in on-treatment compared to pre-treatment biopsies, even in HPV-negative and PD-L1 low tumors. Favorable spatial interactions between DC-1, CD8, CD4, and B cells was associated with PFS and disease control. Conclusions: Results suggest that combination therapy induced formation of highly organized lymphoid aggregates in HNSCC tumors, with a high density of activated B cells, DC and T cells. Together with similar observations indicating that combination immunotherapy with pepinemab induces mature lymphoid structures in tumors of patients with metastatic melanoma, provides evidence of treatment-induced biologic activity corresponding with disease control and suggests a novel and independent mechanism of pepinemab to enhance immune interactions and ICI activity. Clinical trial information: NCT04815720 .