Antibody-drug Conjugate Targeting Research Articles

Expression of Potential Antibody-Drug Conjugate Targets in Cervical Cancer.

(1) Background: There is a huge unmet clinical need for novel treatment strategies in advanced and recurrent cervical cancer. Several cell membrane-bound molecules are up-regulated in cancer cells as compared to normal tissue and have revived interest with the introduction of antibody-drug conjugates (ADCs). (2) Methods: In this study, we characterize the expression of 10 potential ADC targets, TROP2, mesotheline, CEACAM5, DLL3, folate receptor alpha, guanylatcyclase, glycoprotein NMB, CD56, CD70 and CD138, on the gene expression level. Of these, the three ADC targets TROP2, CEACAM5 and CD138 were further analyzed on the protein level. (3) Results: TROP2 shows expression in 98.5% (66/67) of cervical cancer samples. CEACAM5 shows a stable gene expression profile and overall, 68.7% (46/67) of cervical cancer samples are CEACAM-positive with 34.3% (23/67) of cervical cancer samples showing at least moderate or high expression. Overall, 73.1% (49/67) of cervical cancer samples are CD138-positive with 38.8% (26/67) of cervical cancer samples showing at least moderate or high expression. (4) Conclusions: TROP2, CEACAM5 or CD138 do seem suitable for further clinical research and the data presented here might be used to guide further clinical trials with ADCs in advanced and recurrent cervical cancer patients.

A novel SLC3A2-targeting antibody-drug conjugate exerts potent antitumor efficacy in head and neck squamous cell cancer

The development of innovative therapeutic strategies for head and neck squamous cell carcinoma (HNSCC) is a critical medical requirement. Antibody-drug conjugates (ADC) targeting tumor-specific surface antigens have demonstrated clinical effectiveness in treating hematologic and solid malignancies. Our investigation revealed high expression levels of SLC3A2 in HNSCC tissue and cell lines. This study aimed to develop a novel anti-SLC3A2 ADC and assess its antitumor effects on HNSCC both in vitro and in vivo. This study developed a potent anti-SLC3A2 ADC (19G4-MMAE) and systematically investigated its drug delivery potential and antitumor efficacy in preclinical models. This study revealed that 19G4-MMAE exhibited specific binding to SLC3A2 and effectively targeted lysosomes. Moreover, 19G4-MMAE induced a significant accumulation of reactive oxygen species (ROS) and apoptosis in SLC3A2-positive HNSCC cells. The compound demonstrated potent antitumor effects derived from MMAE against SLC3A2-expressing HNSCC in preclinical models, displaying a favorable safety profile. These findings suggest that targeting SLC3A2 with an anti-SLC3A2 ADC could be a promising therapeutic approach for treating HNSCC patients.

Abstract PO3-14-08: Metaplastic Breast Cancer: Whole Genome Sequencing Analysis and Emerging Therapeutic Targets

Abstract Background: Metaplastic breast cancer (MpBC) is a rare histologic subtype of triple negative breast cancer (TNBC) with an aggressive behavior and poor outcome. Previous studies by whole-exome and targeted sequencing showed that MpBCs display a dominant mutational signature 3, associated to homologous recombination (HR) deficiency (HRD). Tumors displaying HRD are associated with response to therapies causing double strand breaks. Nonetheless, MpBCs consistently display significantly lower rates of response to neoadjuvant chemotherapy than non-metaplastic TNBCs and effective therapeutic alternatives are an unmet clinical need. Here we sought to assess the HR features of MpBCs using whole-genome sequencing (WGS), and to evaluate the expression in MpBCs of antigens that are targets of novel antibody drug conjugates (ADCs) in common forms of breast cancer. In addition, we sought to assess the efficacy of ADCs and drugs targeting the DNA damage response (DDR) using preclinical MpBC models. Materials and Methods: We subjected 25 primary MpBC to WGS. HRDetect, the ‘gold standard’ tool for assessment of HRD, was utilized. Mutational signatures were inferred using Signal. 31 MpBCs were subjected to RNA-sequencing to assess the gene expression levels ADC targets compared to non-MpBCs from The Cancer Genome Atlas (TCGA). Efficiency of ADCs and DDR drugs was evaluated using the MpBC cell models HCC1806, BT549 and Hs578T in vitroand/or using HCC1806 xenografts in athymic mice in vivo. Results: Although the majority of MpBCs (13/25) displayed a dominant SBS mutational signature 3, only a small subset of cases (3/25) were classified as HRD by HRDetect. These three bona fide HRD cases were found to harbor bi-allelic inactivation of BRCA1 (n=2) or RAD51C (n=1), and displayed various genomic features of HRD, including deletions with microhomology, enrichment in tandem duplications, rearrangement signatures 3 and 5, and high large scale state transitions (LST) and telomeric allelic imbalance (NtAI) scores. Most MpBCs (22/25), however, were not classified as HRD by HRDetect displaying an incomplete HRD phenotype, with only partial genomic features of HRD. 52% of MpBCs (13/25) displayed APOBEC mutagenesis exposure, which was present in the three bona fide HRD MpBCs. RNA-sequencing analyses revealed detectable ERBB2 and TROP2 expression in all MpBCs interrogated, although at lower levels than in non-MpBCs from TCGA. The HCC1806, BT549 and Hs578T MpBC cell models were found to be sensitive to Sacituzumab Govitecan and Trastuzumab Deruxtecan (T-DXd) in cell viability and colony formation assays, and to T-DXd in vivo. These MpBC cell models were also found to be sensitive to DDR compounds, such as ATR, WEE1 and PKMYT1 inhibitors in vitro, alone and in combination with T-DXd or Sacituzumab Govitecan. Conclusions: Most MpBCs display an incomplete HRD genomic scar, except for those cases harboring bi-allelic inactivation of HRD genes. APOBEC mutagenesis is operative in MpBCs and co-exists with HRD. MpBCs express ERBB2 and TROP2, and MpBC cell models are sensitive to T-DXd and Sacituzumab govitecan, as well as DDR compounds, targeting the incomplete HRD phenotype of MpBCs, alone and in combination. Taken together, these findings warrant further studies in patients with MpBCs testing the therapeutic efficacy of agents targeting the incomplete HRD phenotype, as well as the expression of ADC targets in these aggressive rare forms of TNBCs. Citation Format: Fresia Pareja, Andrea Gazzo, Higinio Dopeso, David Brown, Pier Selenica, Yingjie Zhu, Juan Blanco-Heredia, Laxmi Gusain, Xin Pei, Giacomo Montagna, Nour Abuhadra, Hanna Y Wen, Edi Brogi, Nadeem Riaz, Sarat Chandarlapaty, Maurizio Scaltriti, Larry Norton, Simon Powell, Jorge Reis-Filho, Britta Weigelt. Metaplastic Breast Cancer: Whole Genome Sequencing Analysis and Emerging Therapeutic Targets [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-14-08.

Abstract PO1-14-04: A multivariate biomarker to guide antibody-drug conjugate selection and provide insight on response differences across breast cancer subtypes

Abstract Background: Antibody-drug conjugates (ADCs), including trastuzumab deruxtecan (T-DXd; targeting HER2) and sacituzumab govitecan (SG; targeting TROP2), have transformed the management of metastatic breast cancer, with additional ADCs approved in other solid tumors or in late-stage development. To date, expression of ADC target alone has been poorly predictive of objective response rates (ORRs) in both breast cancer and other tumors. Hence, there is an urgent need to develop better predictive biomarkers to guide ADC vs. other treatments, T-DXd vs. SG treatment, and optimized predictive medicine opportunities involving other ADCs. Additionally, whether ADCs are similarly effective in less common breast cancer subtypes—such as invasive lobular carcinoma (ILC) vs. invasive ductal carcinoma (IDC), is unclear. Given the limited availability of trial tissue samples and cohorts with long-term follow-up, herein we sought to determine whether a tissue based, pan-solid tumor, multivariate biomarker algorithm could predict observed ORRs across ADCs and tumor types, as has been used previously to associate tumor mutation burden with immunotherapy ORRs. Methods: From 15,032 FFPE tumor tissue samples (from 21 tumor types) tested by clinical comprehensive genomic profiling plus RNA based quantitative transcriptional profiling (qTP) as part of the observational Strata Trial (NCT03061305), the ADC treatment response score (TRS) was discovered and validated to predict published ORRs across tumor types and approved/late-stage ADCs (n=16 observed ORRs [from 7 tumor types and 8 ADCs; SG not included]). The best performing 3-factor algorithm (by Pearson correlation coefficient of observed ORRs vs. tumor type and ADC specific predicted biomarker positivity rates) included only qTP components and combined ADC target expression, cell proliferation, and extracellular matrix adhesion (the latter being negatively associated with ORRs). Importantly, predictive biomarker positivity rates of TRS was more correlated vs. observed ORRs (n=16, r=0.81, p=0.0001) than target expression alone (n=16, r=0.54, p=0.03). TRS was then validated using held out SG ORRs from nine tumor types in the IMMU-132-01 basket trial and two ADC ORRs from ASCO 2023 abstracts (enfortumab vedotin [EV; targeting NECTIN-4] in head and neck cancer and patritumab vedotin [targeting HER3] in breast cancer), with TRS predictive biomarker positivity rates again being more significantly correlated vs. observed ORRs than target expression alone (n=11, TRS r=0.91, p=0.0001; target expression alone r=44, p=0.17). Lastly, the locked TRS model was then applied to the DESTINY-PanTumor02 T-DXd dataset presented at ASCO 2023 (n=21 tumor type/HER2 expression groups matched to the trial groups), with TRS predictive biomarker positivity rates again being highly correlated vs. observed ORRs (n=21, TRS r=0.80, p< 0.0001). Across the 21 tumor types in the Strata Trial dataset, breast cancer had the highest percentage (76%) of patients predicted positive for at least one ADC, with 23% and 55% positive for the approved ADCs mirvetuximab soravtansine (targeting FOLR1) and EV, respectively. Lastly, patients with ILC vs. IDC had significantly greater TRS positivity rates for T-DXd (61% vs. 47%) and SG (57% vs. 48%) due to significantly decreased extracellular matrix adhesion in ILC vs. IDC. Clinical outcomes data for patients treated with ADCs and available TRS enrolled in the Strata Trial are maturing and will be presented at the meeting. Conclusion: We have developed and validated TRS, a multivariate RNA based tumor tissue algorithm that predicts observed ORRs across tumor types and approved/late-stage ADCs. More than 75% of all patients with metastatic breast cancer are predicted to be responsive to one or more ADCs, including those approved in other tumor types. Citation Format: Laura Lamb, Jonathan Mowers, Azadeh Nasrazadani, Mark Burkard, Nickolay Khazanov, Daniel Hovelson, Kat Kwiatkowski, Scott Tomlins, D. Bryan Johnson, Daniel Rhodes. A multivariate biomarker to guide antibody-drug conjugate selection and provide insight on response differences across breast cancer subtypes [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-14-04.

Abstract PO4-15-10: HER2-low Status among Patients with Li-Fraumeni Syndrome and Breast Cancer

Abstract Background: The HER2-low status has gained significance as a target for HER2-directed antibody drug conjugates in breast cancer. It is observed in approximately 60% of hormone receptor-positive (HR+) tumors and 30% of hormone receptor-negative (HR-) tumors. Some studies suggest that the HER2-low status by itself does not influence tumor biology or prognosis. Breast cancer in the context of Li-Fraumeni (LFS) syndrome has been characterized by an enrichment in HER2-positive (HER2+, defined as +3 in the immunohistochemistry [IHQ] or +2 in the IHQ with positive in situ hybridization [ISH]) disease. Considering this, we aimed to evaluate whether LFS patients with breast cancer also exhibit a higher frequency of HER2-low status. Methods: This retrospective-prospective cohort study included patients diagnosed with breast cancer who had a pathogenic/likely pathogenic TP53 germline variant and were treated at two cancer institutions between 1999 and 2023. The primary objectives were to determine the overall proportion of HER2-low status and its distribution according to hormone receptor status. HER2-low was defined as IHQ +1 or IHQ +2 with negative ISH. Results: Fifty-three patients (52 female and 1 male) were included in the study. The most common germline TP53 variant was the TP53 R337H (71.7%). The median age at breast cancer diagnosis was 39 years (range 21–62). The majority of patients had breast cancer of no special type (79%) or lobular carcinoma (11.3%), grade 2 (54.7%) or grade 3 (24.5%), and stage I-II (67.9%). Breast cancer subtypes would classically be classified as HER2+ in 34%, HR+HER2- in 58.5%, and triple-negative breast cancer (TNBC) in 7.5% of the patients. Biopsy samples were obtained from the primary tumor in 96.3% of the cases and from lymph nodes in 3.7%. Overall, 15.1% of the cases exhibited HER2-low status, 34% were HER2-negative, and 17% were HER2-negative without IHC details (Table). In a hypothetical scenario assuming all cases without IHC details were HER2-low, the maximum estimated frequency of HER2-low status would be 32.1%. Among HR+ patients, 25.8% had HER2-low status, 45.2% were HER2-negative, and 29% were HER2-negative without IHC details. All four patients with TNBC exhibited HER2-zero status. Conclusion: Contrary to our initial hypothesis, the frequency of HER2-low status in this cohort of LFS patients with breast cancer was lower than anticipated. These results suggest that while breast cancer developing in the context of germline TP53 variants appears to be associated with HER2 amplification, it does not impact lower expressions of HER2. This observation reinforces the concept that HER2-low status does not seem to play a significant role as a driver of tumorigenesis. Further expansion of the study cohort and pathology analysis is planned to confirm these findings. Table HER2 status overall and according to hormone receptor-status (for HER2-negative patients). Abbreviations: IHQ, immunohistochemistry; HR+HER2- BC, hormone receptor-positive HER2-negative breast cancer; TNBC, triple-negative breast cancer. Citation Format: Luciana Auresco, Renata Colombo Bonadio, Vanessa Petry, Laura Testa, Camila Moniz, Jennifer Dos Santos, Maria Del Pilar Estevez-Diz, Rodrigo Santa Cruz Guindalini. HER2-low Status among Patients with Li-Fraumeni Syndrome and Breast Cancer [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO4-15-10.

Abstract PS11-06: Analysis of HER2 expression changes from breast primary to brain metastases including HER2 Low and impact on overall survival

Abstract Background: Previous studies have shown that breast cancer receptor subtype switching between matched primary and metastatic brain metastases (BrM) is common. HER2 expression at low levels, termed “HER2 Low”, has emerged as a new therapeutic biomarker for highly active antibody drug conjugates with potential intracranial activity. Trastuzumab deruxtecan is FDA approved for the treatment of HER2 Low and HER2 positive breast cancer with emerging evidence of CNS responses and clinical benefit in patients with BrM. This study aimed to investigate the relationship between HER2 expression including HER2-Low between matched breast cancer primary tumors and brain metastases. We also sought to evaluate the impact of low HER2 expression on overall survival in breast cancer brain metastases. Methods: We identified pts with MBC seen at least once at a single, NCI-designated center and had a diagnosis of BrM between 2003 and 2023. Of 1252 MBC pts with a diagnosis of BrM, 265 pts underwent resection/biopsy for BrM with available ER/PR/HER2 status. Only pts with available matched primary and BrM tissue were analyzed. HER2 expression was defined as: HER2+ (3+ or 2+/ISH amplified), HER2-Low (1+, 2+/ISH neg) or HER2-0 [by ASCO-CAP guidelines]. Estrogen receptor (ER) status was defined as ER≥1%. Multivariate overall survival (OS) analyses by Cox proportional hazard models were determined from time of BrM resection to death or last follow-up between HER2+, HER2-Low and HER2-0 patients controlling for ER and age. Results: 197 matched breast cancer primary and BrMs with available ER, PR, and HER2 testing were evaluated. Median age at BrM resection was 51 yrs. ER was positive in 48% (N=95) of BrMs. Of 265 resected BrMs, HER2 expression was found in 72%: 49.8% HER2+ (N=132), 22.2% HER2-Low (N=59), and 27.9% HER2-0 (N=74). HER2 expression by 197 primary tumors was: 57% HER2+ (N=112), 24% HER2-Low (N=48), 19% HER2 0 (N=37). Amongst 112 HER2+ primary tumors, 97% (N=109) had concordant HER2+ BrMs, 3 (2.6%) switched to HER2-Low, with 100% retaining some HER2 expression. Of 37 HER2-0 primaries, 35% (13/37) gained HER2-Low expression and 5.4% (2/37) gained HER2+ expression in the BrM. Amongst 48 HER2-Low primary tumors, half (52%) changed expression to either HER2+ (21%) or HER2-0 (31.2%) in the brain. 66 tumors had extracranial tissue biopsied prior to BrM resection available for HER2 analysis. Of 15 HER2-0 extracranial tumors, 47% gained HER2-Low and 13% gained HER2+ at time of BrMs. After adjusting for age and ER status at the time of BrMs, patients with HER2+ BrM had a statistically significant lower risk of death at time of follow up than those with HER2-Low BrM (HR=0.41, P=0.0006). The risk of death between patients with HER2-0 BrM did not differ from HER2-Low BrM after adjustment for ER and age (HR=0.96, p =0.9). Median survival from BrM resection was 12.4m, 16.1m, 47.6m for HER2-Low, HER2-0 and HER2+ respectively (p = .0004). Conclusions: HER2 expression amongst primary, extracranial and brain metastases is dynamic with frequent change. Over 40% of HER2-0 primary or extracranial tumors gain HER2 expression in the brain. These changes have therapeutic implications given HER2 targeting antibody drug conjugates with improved OS and emerging evidence of CNS activity, highlighting the need for better diagnostics, including those that do not require invasive tissue sampling, to determine HER2 status in the brain. Patients with HER2-Low and HER2-0 BrMs have inferior survival and are a clinical unmet need. Citation Format: Alyssa Pereslete, Melissa Hughes, Alyssa Patterson, Janet Files, Kyleen Nguyen, Lauren Buckley, Ashka Patel, Abigail Moore, Eric Winer, Tianyu Li, Sara Tolaney, Nancy Lin, Sarah Sammons. Analysis of HER2 expression changes from breast primary to brain metastases including HER2 Low and impact on overall survival [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PS11-06.

Abstract PO1-26-02: Enfortumab vedotin (EV): Correlation of estrogen receptor status and Nectin-4 expression with single agent efficacy in breast XPDX models

Abstract EV is a Nectin-4 targeting antibody-drug conjugate (ADC) with an MMAE payload, recently approved for treatment of bladder cancer patients. To better understand the potential for EV in treatment of breast cancer, we evaluated 175 breast XPDX models, including hormone receptor positive and negative cancer, some with actionable mutations, representing primary and metastatic disease from naïve or clinically treated patients. Each model was stained and scored for Nectin-4 protein and evaluated in vivo against single agent EV and activity correlated with ER status, Nectin-4 staining and known variants. 175 breast XPDX models were evaluated in this study, approximately 35% representing estrogen receptor positive and 20% clinically HER2+ cancers. Nectin-4 protein expression based on intensity and proportion was determined on a scale of 0+-3+, and models profiled using WES and RNAseq. For in vivo studies, models were evaluated against single agent EV administered by intravenous injection once weekly for three cycles at 3 mg/kg. Endpoints included tumor volume and time from treatment initiation with %T/C values and tumor regression reported at study completion; a T/C of ≤ 20% versus control was considered sensitive. Tumor regression (%T/C< 0%) versus Day 0 tumor volume was also reported. Overall, 40% of breast models stained positive for Nectin-4 including 50% of ER- and 20% of ER+ and 50% of HER2+ models. For ER- models, 50% were negative and 40% stained 1+ while 80% of ER+ models were negative and 15% stained 1+. In vivo, 25% of models were sensitive to EV, with 65% of these models ER- and < 10% HER2+. Nectin-4 staining did not directly correlate to model sensitivity although models with known drivers such as AKT1 and ESR1 were insensitive as were several models established from post CDK4/6i patients. We have characterized a panel of breast XPDX models based on Nectin-4 staining and in vivo sensitivity to EV and correlated activity with protein and receptor staining and known mutations. This data is a valuable tool in further developing EV and identifying its potential in treating breast cancer. Citation Format: Crystal Moreno, Johnnie Flores, Alyssa Simonson, Maci DeBoer, Natalia Banos, Armando Diaz III, Morgan Lynch, Jim Lund, Leonora Leykum, Tahmi Rouzbahan, Manuel Pedregal, Kyriakos P. Papadopoulos, Amy Lang, Arthur Rosenthal, Drew Rasco, Gladys Rodriguez, Luis Rodriguez, Nehal Lakhani, Muralidhar Beeram, Amita Patnaik, Maria De Miguel, Ronald Drengler, Steven Abbate, Bernard Doger de Speville Uribe, Michael Wick. Enfortumab vedotin (EV): Correlation of estrogen receptor status and Nectin-4 expression with single agent efficacy in breast XPDX models [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-26-02.

Preparation and evaluation of the PD0721‑DOX antibody‑drug conjugate targeting EGFRvIII to inhibit glioblastoma

Preparation and evaluation of the PD0721‑DOX antibody‑drug conjugate targeting EGFRvIII to inhibit glioblastoma

A novel bispecific antibody drug conjugate targeting HER2 and HER3 with potent therapeutic efficacy against breast cancer.

Antibody drug conjugate (ADC) therapy has become one of the most promising approaches in cancer immunotherapy. Bispecific targeting could enhance the efficacy and safety of ADC by improving its specificity, affinity and internalization. In this study we constructed a HER2/HER3-targeting bispecific ADC (BsADC) and characterized its physiochemical properties, target specificity and internalization in vitro, and assessed its anti-tumor activities in breast cancer cell lines and in animal models. The HER2/HER3-targeting BsADC had a drug to antibody ratio (DAR) of 2.89, displayed a high selectivity against the target JIMT-1 breast cancer cells in vitro, as well as a slightly higher level of internalization than HER2- or HER3-monospecific ADCs. More importantly, the bispecific ADC potently inhibited the viability of MCF7, JIMT-1, BT474, BxPC-3 and SKOV-3 cancer cells in vitro. In JIMT-1 breast cancer xenograft mice, a single injection of bispecific ADC (3 mg/kg, i.v.) significantly inhibited the tumor growth with an efficacy comparable to that caused by combined injection of HER2 and HER3-monospecific ADCs (3 mg/kg for each). Our study demonstrates that the bispecific ADC concept can be applied to development of more potent new cancer therapeutics than the monospecific ADCs.

Abstract LB043: VBC101-F11: An innovative EGFR/cMet bispecific antibody drug conjugate (ADC) targeting key oncogenic drivers in solid tumors

Abstract EGFR and cMet are two classic growth factor receptors frequently co-expressed in solid tumors at significantly high levels, while only limited to moderate expression in normal tissues. Beyond their co-expression patten, the aberrant or over expression in their pathway are implicated in oncogenesis. Notably, MET pathway activation either through increased cMet expression or its ligand, is often associated with resistance to EGFR tyrosine kinase inhibitors (TKI). Therefore, the interplay between EGFR and cMet, coupled with their differential expression in tumors versus normal tissue substantiates the rationale for a bispecific approach that targets both receptors to enhance tumor specificity. VBC101-F11 is a bispecific anti-EGFR/cMet ADC engineered to deliver clinically validated cytotoxic agents to tumor cells with a drug-to antibody ratio (DAR) of 4. This innovative design distinguishes VBC101-F11 from other EGFR/cMet-targeting agents on the market or in clinical trials (ABBV399 or AZD9592), optimizing both efficacy and safety. 1. VBC101-F11 features a low-affinity EGFR arm coupled with a high-affinity cMet arm. Incorporating the low-affinity EGFR arm enhances binding avidity and internalization, resulting in a superior cytotoxicity with a 5-fold improvement in IC50 compared to ABBV-399 (targeting cMet only). This nuanced EGFR arm design ensures a reduction in EGFR-related off-target toxicity in normal tissues. 2. The biparatopic design in cMet arm further underscores our strategic rationale by demonstrating a synergistic effect in terms of binding, internalization, and functional blocking with a 10-fold improvement in EC50 compared to the monospecific parental antibodies that target a single epitope. This ensures VBC101-F11’s maximized efficacy in tumor cells. 3. VBC101-F11’s innovative nanobody-based format sets our molecule apart. Living imaging data in mouse model reveals that VBC101-F11 (90kD) achieves superior tumor penetration and accumulation from 1 hour to over 90 hours surpassing the mAb ABBV399 and bispecific antibody AZD9592, both at 150kD. 4. VBC101-F11 has demonstrated outstanding molecular developability in stress tests, particularly in human and non-human primate (NHP) plasma, where both the antibody component and the entire ADC molecule exhibit remarkable stability, positioning VBC101-F11 as a viable candidate for future Chemistry, Manufacturing, and Controls (CMC) development. 5. In in vivo CDX models using various lung cancer cell lines, VBC101-F11 equipped with MMAE warhead displayed a marked superiority in tumor growth inhibition (TGI of 60% vs. 17%) at a single dose of 3mg/kg compared to MMAE conjugating ADC ABBV-399. Further payload exploration, such as DNA replication inhibitor, has also yielded promising results, with VBC101-F11-new payload showing comparable efficacy to the same class ADC AZD9592 in models with low EGFR and high cMet expression (100% TGI) and in those with moderate EGFR/cMet expression (~ 60% TGI). In summary, VBC101-F11, with unique design and distinguished attributes, has demonstrated its remarkable potential in the landscape of anti-EGFR/cMet therapies. The encouraging efficacy observed in preclinical models supports the molecule’s advancement into clinical development, positioning it as a best-in-class bispecific ADC for treating tumors in patients. Citation Format: Wei (Vivian) Wang, Jing Li, Lin Ma, Man Xu, Qun Yin. VBC101-F11: An innovative EGFR/cMet bispecific antibody drug conjugate (ADC) targeting key oncogenic drivers in solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB043.

Abstract CT155: Trial in progress: A phase Ia/Ib, an open-label, multicenter study of ZL-1310 to evaluate the safety, tolerability, and pharmacokinetics in subjects with small cell lung cancer

Abstract Background: ZL-1310, a novel antibody-drug conjugate (ADC) targeting delta-like protein 3 (DLL3) with a camptothecin derivative as its payload via a protease-cleavable linker, employs TMALIN® (Tumor Microenvironment Activable LINker-payload) ADC technology platform. DLL3, an inhibitory Notch pathway ligand and a validated target for directed therapy, is highly expressed in more than 80% of patients with small cell lung cancer (SCLC) and also has variable expression in other neuroendocrine (NE) tumors. There remains an unmet medical need for 2nd-line SCLC, as the standard of care chemotherapy has a modest response rate of ~25%. ZL-1310 is designed to minimize payload toxicity and improve anti-tumor effectiveness. Methods: This is an open-label, multiple-dose, phase 1 study of ZL-1310 administered to subjects with relapsed/refractory (r/r) metastatic SCLC who have progressed after at least one platinum-based chemotherapy regimen. The study consists of two parts: Part 1 (dose escalation) and Part 2 (dose expansion). In Part 1, the Bayesian optimal interval (BOIN) design is employed for dose escalation, while in Part 2, patients will be randomized into two cohorts with different doses of ZL-1310 to further define safety and preliminary antitumor activity. The major inclusion criteria are as follows: (1) Subjects must have histologically or cytologically confirmed metastatic or extensive-stage SCLC with no more than 3 prior regimens in the r/r setting. (2) Subjects must have at least one measurable target lesion as defined by RECIST v1.1. (3) Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1. The major exclusion criteria include: (1) Prior exposure to DLL3-targeted therapy. (2) Subjects with symptomatic or uncontrolled brain metastasis requiring concurrent treatment, impaired major organ functions, active autoimmune disease, or active infections.Part 1 Dose Escalation is recruiting subjects in the US and China. Citation Format: Afshin Dowlati, Alexander Spira, Xiaodong Shen, Yinjia Fu, Yiyuan Pan, Linda Liu, Renke Zhou, Yi-Long Wu. Trial in progress: A phase Ia/Ib, an open-label, multicenter study of ZL-1310 to evaluate the safety, tolerability, and pharmacokinetics in subjects with small cell lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr CT155.

Abstract LB025: CKD-703, a novel antibody-drug conjugate targeting cMET with an enhanced therapeutic index in solid tumors

Abstract In our pursuit of advancing cancer therapeutics, we introduce CKD-703, a groundbreaking cMET-targeting antibody drug conjugate (ADC) designed to strategically address the challenges posed by cMET overexpression and alteration in various solid tumors. Recognizing the pivotal role of tolerability and efficacy in shaping the therapeutic landscape of ADC, CKD-703 represents a significant advancement, aiming to widen the therapeutic index of cMET-targeted therapies. As we develop CKD-703, an innovative cMET-targeting ADC, we acknowledge the oncogenic potential of cMET alterations, particularly in non-small cell lung cancer (NSCLC), and the prevalent elevation of cMET across various cancers. Antibody drug conjugates targeting cMET (cMET-ADCs) have shown its therapeutic potential in treating cMET positive tumors, irrespective of their dependency on cMET signaling. Clinical success, however, has often been limited to NSCLC patients with the highest cMET levels. In light of this, CKD-703 is crafted to not only improve tolerability but also enhance efficacy, thereby broadening its application to a diverse patient population, including those with moderate cMET expression. To develop a novel cMET-ADC, Chong Kun Dang leverages Synaffix's ADC platform technology, incorporating glycan remodeling to achieve site-specific conjugation of the linker-payload to the antibody and ensuring a homogeneous drug-antibody ratio. These unique characteristics enhance CKD-703's stability in the bloodstream, minimizing the concentration of liberated free payload and significantly improving tolerability. The uniform DAR of CKD-703 contributes to improved pharmacokinetics, ensuring a consistent and sustained therapeutic effect. In comparison to benchmark antibodies, CKD-703 exhibits superior binding affinity to cMET and efficient internalization, resulting in enhanced therapeutic efficacy. This is evidenced in preclinical experiments which performed in several animal models that can mimic the complexities of cMET positive tumors with varying expression levels. The GLP toxicity studies of CKD-703 are ongoing. In conclusion, CKD-703 is a novel cMET-targeting ADC designed to overcome the limitations observed in existing cMET-ADCs in clinical trials as well as significantly improve tolerability and efficacy. By strategically addressing challenges associated with moderate cMET expression levels, CKD-703 emerges as a promising therapeutic candidate, aiming to broaden the therapeutic index in cMET-overexpressing solid tumors. This study highlights the importance of tailored approaches to diverse patient populations in advancing the development of CKD-703 in the pursuit of precision cancer therapies. Citation Format: Kwanwoo Lee, Sue-Youn Kim, Hye-Jin Hong, Eunjeong Park, Jungok Park, Seong-Ho Hong, Daekwon Bae, Nina Ha, Sohn-Ji Yeon, Inchang Hwang, Semi Kim, Yoonseok Lee, Ju-Hee Lee. CKD-703, a novel antibody-drug conjugate targeting cMET with an enhanced therapeutic index in solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB025.

Abstract CT084: A phase 1 study of LY4101174 (ETx-22), an antibody-drug conjugate targeting nectin-4, in patients with advanced or metastatic urothelial cancer and other solid tumors (trial in progress)

Abstract Background: Nectin-4 is a cell adhesion molecule expressed in multiple tumor types.1 Enfortumab vedotin (EV), an antibody-drug conjugate (ADC) comprising a fully humanized nectin-4 antibody conjugated to the microtubule inhibitor, monomethyl auristatin E (MMAE), is approved for the treatment of advanced urothelial cancer (UC). However, skin toxicity, peripheral neuropathy, and hyperglycemia are clinically significant toxicities with EV. LY4101174 (ETx-22) is a next-generation nectin-4 targeting ADC comprising a humanized IgG1 Fc-silent monoclonal antibody linked to the topoisomerase I inhibitor, exatecan, via a maleimide-ß-glucuronide poly-sarcosine linker with improved stability and a homogeneous drug-antibody ratio (DAR) of 8. Preclinically, LY4101174 has demonstrated robust in vivo efficacy in tumor models across a range of nectin-4 expression levels and in the nectin-4-MMAE ADC treated resistant tumor model.2 Methods: LOXO-ENC-23001 is a global, open-label, multi-center, first-in-human phase 1 study of LY4101174 in patients (pts) with advanced or metastatic solid tumors known to express nectin-4. LY4101174 will be administered intravenously and evaluated initially on either every 2 weeks or 3 weeks schedule. Eligible pts (≥18 years) must have good performance status (ECOG PS 0-1) with metastatic UC, head and neck squamous cell carcinoma, esophageal, pancreatic, prostate, non-small cell lung, cervical, ovarian, or triple-negative breast cancer, and must have received or not be a candidate for available standard therapies. The study will be conducted in 2 phases: phase 1a dose escalation/dose optimization and phase 1b dose expansion. Dose escalation will utilize a Bayesian optimal interval design. Additional pts will be allowed to backfill to previously cleared dose levels that demonstrate therapeutically relevant exposures or direct evidence of clinical activity. The dose-limiting toxicity evaluation period will be 21 or 28 days based on dosing schedules. Phase 1b dose expansion cohorts will enroll pts with selected tumor types: mUC with no prior EV treatment (cohort B1), mUC with prior EV treatment (cohort B2), and other selected non-UC tumors (cohort C). Primary objectives are to determine the recommended phase 2 dose (RP2D) and assess the safety of LY4101174. Key secondary objectives are to evaluate the pharmacokinetic profile, immunogenicity, and antitumor activity of LY4101174 per RECIST v1.1. Nectin-4 expression and other biomarker data will be generated and correlated with clinical activity. 1Challita-Eid PM et al. 2016 Cancer Res 76(10):3003-13. 2Azim H et al. 2023 Mol Cancer Ther 22(12_Suppl):B128. Citation Format: Jonathan Rosenberg, Renaud Sabatier, Armelle Viceneux, Thibault de La Motte Rouge, Stephane Champiat, Loic Lebellec, Philippe Barthélémy, Guru Sonpavde, Xin Gao, Scot Niglio, Tian Zhang, Craig Gedye, Matthew Galsky, Matthew Milowsky, Melissa Reimers, Andrew Weickhardt, Enrique Grande, Daniel E. Castellano, Mehmet Bilen, Amita Patnaik, Robert Zerbib, Minna Balbas, Xiaojian Xu, Kaitlyn Aptaker, Arjun V. Balar, Ahmad H. Awada. A phase 1 study of LY4101174 (ETx-22), an antibody-drug conjugate targeting nectin-4, in patients with advanced or metastatic urothelial cancer and other solid tumors (trial in progress) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr CT084.

Abstract LB127: A first-in-class bispecific ADC targeting FORL1 and MUC1 exhibits promising anti-tumor activity in a FOLR1low xenograft model

Abstract Folate receptor alpha (FOLR1) is specifically overexpressed in malignant tumors of epithelial origin, including ovarian, lung and breast cancers. While antibody-drug conjugate (ADC) therapies targeting FOLR1 have shown promise in clinical trials, there is still a large proportion of patients, particularly those with low FOLR1 expression, who do not respond, indicating new therapeutic strategies are needed. MUC1 is another tumor-associated antigen that is likewise overexpressed in epithelial ovarian cancer, breast cancer, and NSCLC, among other cancers; analysis of ovarian PDX tumors by immunohistochemistry (IHC) suggests that FOLR1 and MUC1 are both highly expressed in this cancer type. We therefore hypothesized that dual targeting of FOLR1 and MUC1 could provide therapeutic benefit in ovarian cancer as well as a broad range of tumors. We generated a FOLR1 x MUC1 bispecific antibody (bsAb) from anti-FOLR1 and anti-MUC1 monoclonal antibodies that were produced using RenLite® fully human common light chain mice. These parental monoclonal antibodies recognize both human and cynomolgus monkey antigens. Anti-FOLR1 antibodies were highly specific for FOLR1, as they did not recognize other FOLR family members. The anti-MUC1 antibody was previously confirmed to target MUC1-C, the juxtamembrane domain of MUC1, ensuring tumor cell recognition regardless of MUC1 cleavage. Assessment of the novel FORL1 x MUC1 bsAb indicates binding to several ovarian, breast and NSCLC cell lines, and shows enhanced internalization in MDA-MB-468 and T47D cell lines when compared to parental antibodies. Subsequently, the FORL1 x MUC1 bsAb was conjugated to vcMMAE with a DAR of 4, to generate a bispecific ADC, BCG023. BCG023 exhibited superior tumor growth inhibition compared to benchmark ADCs conjugated with the same payload and DAR in a low FOLR1-expressing NSCLC PDX model. These data suggest that targeting FOLR1 and MUC1 with a bispecific ADC is a promising therapeutic strategy for the treatment of ovarian and other tumors. Citation Format: Yifu Zhang, Chengzhang Shang, Chaoshe Guo, Yi Yang. A first-in-class bispecific ADC targeting FORL1 and MUC1 exhibits promising anti-tumor activity in a FOLR1low xenograft model [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB127.

Durable Response to Enfortumab Vedotin Compared to Re-challenging Chemotherapy in Metastatic Urothelial Carcinoma After Checkpoint Inhibitors.

Enfortumab vedotin (EV), an antibody-drug conjugate targeting Nectin-4, has been used for patients with metastatic urothelial carcinoma (mUC) after progressing on checkpoint inhibitors (CPIs). Re-challenging chemotherapy with platinum agents and continuing CPIs beyond progressive disease (PD) have often been chosen following PD on CPIs, and several studies indicate favorable treatment effects of re-challenging chemotherapy. There is little evidence for comparing EV and re-challenging chemotherapy in real-world clinical practice. The aim was to reveal the real-world treatment outcomes of EV, re-challenging chemotherapy, and continuing CPIs beyond PD in mUC patients. A multi-institutional dataset of 350 mUC patients treated with CPIs was utilized. Overall survival (OS), progression-free survival (PFS), objective response rate (ORR), and duration of response (DOR) were evaluated to compare the treatment arms. One hundred and nine mUC patients were treated with EV with a median follow-up of 6.4months. The ORR and disease control rate (DCR) were 48% and 70%, respectively. The OS from PD on pembrolizumab exhibited significant differences among the three groups, with a median OS of 8, 14, and 29months in continuing pembrolizumab beyond PD, re-challenging chemotherapy, and EV, respectively. When comparing the survival outcomes from the initiation of the treatment, there is neither a difference in OS (p=0.124), PFS (p=0.936), nor ORR (p=0.816) between EV and re-challenging chemotherapy. Notably, the DOR in patients who achieved an objective response was significantly longer in the EV group than the re-challenging chemotherapy group (a median of 11 and 5months, p=0.049). For OS, the difference was not statistically significant (27 and 11months in EV and re-challenging chemotherapy, respectively: p=0.05). A superior effect of EV on patient survival compared to re-challenging chemotherapy and continuing pembrolizumab beyond PD was observed in our real-world analysis, which is attributed to the durable DOR in EV treatment despite the similar ORR to re-challenging chemotherapy.

Therapeutic prospects of nectin-4 in cancer: applications and value.

Nectin-4 is a Ca2+-independent immunoglobulin-like protein that exhibits significantly elevated expression in malignant tumors while maintaining extremely low levels in healthy adult tissues. In recent years, overexpression of Nectin-4 has been implicated in tumor occurrence and development of various cancers, including breast cancer, urothelial cancer, and lung cancer. In 2019, the Food and Drug Administration approved enfortumab vedotin, the first antibody-drug conjugate targeting Nectin-4, for the treatment of urothelial carcinoma. This has emphasized the value of Nectin-4 in tumor targeted therapy and promoted the implementation of more clinical trials of enfortumab vedotin. In addition, many new drugs targeting Nectin-4 for the treatment of malignant tumors have entered clinical trials, with the aim of exploring potential new indications. However, the exact mechanisms by which Nectin-4 affects tumorigenesis and progression are still unclear, and the emergence of drug resistance and treatment-related adverse reactions poses challenges. This article reviews the diagnostic potential, prognostic significance, and molecular role of Nectin-4 in tumors, with a focus on clinical trials in the field of Nectin-4-related tumor treatment and the development of new drugs targeting Nectin-4.

Advances in the treatment of anti-HER-2 antibody drug conjugates in pan-tumor with low HER-2 expression

Antibody-drug conjugates (ADCs) are drugs that combine monoclonal antibody drugs targeting specific antigens and small molecule cytotoxic drugs through linker molecules. ADCs combine the advantages of high specificity targeting and potent killing effects, achieving precise and efficient targeting of cancer cells. Nowadays, ADCs are one of the hotspots in cancer drug development. Human epidermal growth factor receptor 2 (HER-2) is a known oncogene that can drive the occurrence and development of various types of tumors. HER-2 is also an important tumor target for ADCs approved for solid tumors. Anti-HER-2 ADCs can not only be used to treat HER-2-positive tumors but also effectively target HER-2-low tumors. The emergence of ADCs has broken the traditional classification of HER-2 in tumors, bringing significant treatment breakthroughs for HER-2-low tumors. Anti-HER-2 ADCs are widely used in the treatment of solid tumors and have substantial evidence for HER-2-low tumors. This article presents the progress of various anti-HER-2 ADCs in HER-2-low tumors including breast cancer, gastrointestinal malignancies, urothelial carcinoma, lung cancer. And this article summarizes the current status of preclinical studies, clinical studies, and safety of anti-HER-2 ADCs in order to provide reference for the clinical use of HER-2-low tumors.

Abstract 7525: Investigating B7-H3 as a checkpoint and antibody drug conjugate target across prostate cancer variants- both androgen receptor dependent and independent

Abstract Background: CD276 (B7-H3) has recently emerged as a promising presumptive immune checkpoint inhibitor (ICI) and antibody-drug conjugate (ADC) target for PCa. Unfortunately, current immunotherapy trials have yielded few objective responses in PCa and hormonal therapy seldom cures necessitating the need to develop new immunotherapy or targeted antibody-drug conjugate (ADC) treatment options for PCa, including B7-H3 targeted approaches. Method: In this work, we evaluated important ADC targets and immune checkpoints (PD-1, PD-L1, PD-L2, LAG3, TIGIT, OX40, 4-1BB, CTLA4, STEAP1, STEAP2, PSMA, NECTIN1, TROP-2, DLL3 and B7-H3) in various PCa cell lines and multiple patients spanning various subtypes of PCa using publicly available bulk RNA sequencing data and single-cell RNA sequencing (scRNA-seq) data respectively. The results were validated at RNA and protein levels in listed cell lines. Further, the impact of androgen receptor (AR) signaling on B7-H3 expression was quantified using charcoal stripped media, AR agonist (R1881) and antagonist (enzalutamide) using real-time PCR and flow cytometry at RNA and protein levels respectively. Results: We found that B7-H3 shows high expression and very low variability compared to other ADC targets and immune checkpoints in AR-positive (LnCAP), hormone-refractory (VCAP), Castrate resistant (22RVI), AR negative (PC3, DU145), and neuroendocrine (H660) cell lines, as well as in—PCa cells and tumor microenvironment (TME) myeloid cells—of multiple patients spanning various subtypes of PCa. Western blotting confirms the ubiquitous presence of B7-H3 in all these cell lines, even when PSMA is absent. Furthermore, B7-H3 expression was modifiable via androgen depletion, R1881 supplementation, or addition of the AR antagonist enzalutamide. These changes were significant in AR-positive LnCAP cells compared to AR-negative DU145 cells further endorsing the crosstalk between AR and B7-H3 signaling pathways, which will be presented in greater detail. Conclusion: This work explores the therapeutic relevance of important ADC targets and immune checkpoints for Prostate Cancer (PCa), highlighting B7-H3 as a unique therapeutic ADC and ICI candidate across the continuum of PCa—androgen sensitive, castration resistant, and neuroendocrine—due to its stable expression across various subtypes and on TME myeloid cells. The relationship between AR and B7-H3 opens the possibility of future combination therapies based on this interaction. Citation Format: Nikita Mundhara, Shivang Sharma, Emirhan Tekoglu, Ezra G. Baraban, Nathan A. Lack, Eugene Shenderov. Investigating B7-H3 as a checkpoint and antibody drug conjugate target across prostate cancer variants- both androgen receptor dependent and independent [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7525.

Abstract 2911: Prevalence of potential antibody drug conjugate targets and co-expression with PD-L1 in breast cancer and lung cancer patients with clinical response information

Abstract Breast and non-small cell lung cancer (NSCLC) are two of the most important diseases for antibody drug conjugate (ADC) development. ADCs targeting HER2 and TROP2 are currently approved in select breast and lung subgroups with ADC’s to many other tumor targets currently in clinical development. Based on TCGA mRNA expression data, ADC tumor targets currently under investigation for treatment of solid tumors have variable overlap in breast and lung cancers, however little is known about the overlap of ADC targets with PD-L1 which is a predictive biomarker in TNBC and for pembrolizumab monotherapy. In the current study, the prevalence and co-expression of ADC targets (HER2, TROP2, Nectin-4, B7-H3, B7-H4, HER3, FOLR1) were investigated in the context of PD-L1 status and clinical response data for standard-of-care (SOC) and immunotherapy, in breast and NSCLC cancer subtypes. Tissue microarrays (TMAs) comprising treatment naïve tumor samples (1mm cores) and including TNBC (26 patients), ER+ BC (90 patients), and NSCLC (adeno & squamous subtypes; 146 patients), all with full clinical follow-up information, were examined by immunohistochemistry (IHC) in serial sections for each ADC target and PD-L1, applying widely used and well optimized IHC assays on the Ventana and Dako Link 48 platforms. Established scoring methods were employed for analysis of PD-L1 and HER2, except for the addition of an exploratory ultra-low score for HER2. All other ADC targets were scored by digital image analysis to deliver percentage positive tumor expression at high (3+), moderate (2+), weak (1+) and negative (0) intensities. Taken together, this analysis revealed variable prevalence for each target in breast and lung cancer histologies by subtype, including differing overlap between ADC targets within tumor indications. In addition, prevalence of each ADC target in breast cancer characterized by HR status and HER2 subgroups (HER2-positive, low, ultra-low, negative) indicated potential opportunities for ADC clinical development in these subgroups. Co-prevalence of the selected ADC targets with PD-L1 CPS>10 (predictive for Pembrolizumab response in TNBC) and PD-L1 CPS categories of interest for NSCLC revealed patterns of expression in support of combination therapies. Furthermore, expression and co-prevalence data for each ADC target +/- PD-L1 was characterized in association with response to immunotherapy or other SOC. In summary, these data provide better understanding of which breast and lung cancer subtypes express the selected ADC targets uniquely and competitively which may be key to future clinical and diagnostic development. Citation Format: Marie Cumberbatch, Woo Ho Kim, Milan Bhagat, John Cogswell. Prevalence of potential antibody drug conjugate targets and co-expression with PD-L1 in breast cancer and lung cancer patients with clinical response information [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2911.

Abstract 6177: Machine learning enables prediction of ADC targets from whole slide H&E images

Abstract Background: The efficacy of antibody-drug conjugates (ADCs) depends on the expression and specificity of the target antigen by tumor cells. Although H&E-stained slides are routinely collected during cancer care, specialized IHC staining is typically required to ascertain antigen expression. Such stains are not always available or readily deployed. The need to perform separate IHC tests for each candidate ADC may burden clinical labs and can hinder access to care in resource-limited settings. Here we develop an ensemble of machine learning models to accurately predict the expression of 166 distinct ADC targets directly from H&E images. Methods: For each ADC-targeted gene, patients with copy-number amplifications (CNAs) were identified from somatic whole exome sequencing. Genes differentially expressed in patients with CNAs were identified from bulk transcriptomics. For each gene, an expression signature was developed based on the expression levels of differentially upregulated genes. Next, whole-slide, H&E-stained histopathology images were embedded into a lower-dimensional representation via a transformer model trained with self-supervised learning. Neural networks were developed to predict a patient’s probability of having a CNA in an ADC-targeted gene, as indicated by an expression signature exceeding the p90. All evaluation metrics were ascertained by 5-fold cross-validation, with training and evaluation on independent patients. Results: For each of the 166 ADC-targeted genes, a median of 154 patients were found to harbor a CNA, and the expression signature included a median of 180 genes. In all cases, patients with CNAs had significantly higher expression signatures than those without. For predicting likely CNA status directly from H&E histopathology images, the mean AUROC was 0.888 (95% CI, 0.876-0.900) and the mean AUPRC was 0.571 (95% CI, 0.531-0.611). Among the 166 ADC-targeted genes, the AUROC exceeded 0.95 for 31.3%, 0.8 for 80.7%, and 0.725 for 100%. The best-predicted ADC target was SLC7A5 (AUROC: 0.995 [95% CI, 0.994-0.998]; AUPRC: 0.967 [95% CI, 0.963-0.976]). Conclusion: We have developed models that accurately predict the likely expression of ADC targets based solely on H&E images. The ability to accurately discern the presence of ADC antigens from H&E images has numerous potential applications, including cohort refinement, computer-aided diagnosis, and personalized treatment planning. Citation Format: Zachary Ryan McCaw, Anna Shcherbina, Yajas Shah, insitro Research Team, Philip Tagari, Daphne Koller, Christopher Probert. Machine learning enables prediction of ADC targets from whole slide H&E images [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6177.