Search

Background: Sacituzumab govitecan (SG) is an antibody-drug conjugate targeting Trop2 with an SN-38 payload recently approved for pretreated patients with locally advanced or metastatic triple-negative breast cancer (TNBC). The XenoSTART Patient-Derived Xenograft (XPDX) breast cancer platform includes over 180 models spanning all subtypes characterized with immunohistochemistry (IHC) including ER, PR, and HER2 protein levels, genomic and transcriptomic sequencing, and in vivo drug sensitivity. To better understand potential benefit of SG in breast cancers other than TNBC and further annotate our platform, Trop2 protein levels were determined in all breast models by IHC. We evaluated tumor growth inhibition by SG in 125 of our XPDX breast models and compared protein expression with agent activity. Methods: 180 breast XPDX models were evaluated for Trop2 expression (AF650, R&D Systems) and 125 were evaluated in vivo against SG; responses were grouped by ER and Trop2 status (+/-). Models were grown subcutaneously in female athymic nude mice and ER+ models supplemented with estradiol. Models were also characterized for PR, HER2, and AR protein expression by IHC and profiled using WES and RNAseq. For in vivo studies, SG was administered by intravenous injection biweekly for two cycles at 1 mg, flat; 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. Results: 180 breast models were examined by IHC with 75/180 (42%) classified as ER+ and 105/180 (58%) ER-. In ER+ models 38/75 (51%) were Trop2+ and 37/75 (49%) Trop2-, and in ER- models 41/105 (39%) were Trop2+ and 64/105 (61%) Trop2-. In vivo, 20% of ER+/Trop2+ models reported sensitivity to SG, most notably models from patients with acquired resistance to CDK4/6 inhibitors, including STM001 and ST4316B. Interestingly, >70% of ER+/HER2+/Trop2+ models were insensitive to SG, including ST225 and ST340. Of 41 ER-/Trop2+ models, approximately 40% reported some response to SG with 50% of these sensitive to therapy, including ST5954 established from a patient who began treatment with SG following sample collection and is currently in remission. >75% of Trop2- models were insensitive to SG regardless of ER status. Conclusion: We screened 180 models in our XPDX breast cancer platform for Trop2 expression and compared expression with in vivo SG efficacy in 125 models. Analysis is underway to correlate receptor and molecular profiles with SG sensitivity in breast models and we are expanding expression and in vivo testing to additional indications. Citation Format: Alyssa Simonson, Johnnie Flores, Ebony Anderson, Crystal Moreno, George Plasko, Kyriakos P. Papadopoulos, Amita Patnaik, Drew Rasco, Gladys Rodriguez, Amy Lang, Muralidhar Beeram, Luis Rodriguez, Ronald Drengler, Steven Abbate, Hanni Salih, Lon Smith, Maryam Elmi, Brittany DeBerry, Arthur Rosenthal, Tatiana Hernandez, Nehal Lakhani, Manish Sharma, Michael Wick. Correlation of Trop2 expression with in vivo sensitivity to sacituzumab govitecan in a panel of breast XPDX models [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P3-08-05.

Background: Trastuzumab deruxtecan (DS-8201a) is an antibody-drug conjugate (ADC), consisting of a humanized anti-HER2 (human epidermal growth factor receptor 2) monoclonal antibody linked to a topoisomerase I inhibitor payload using a cleavable tetrapeptide-based linker, approved for the treatment of HER2+ metastatic breast cancer patients refractory to anti-HER2 therapy including T-DM1. While some mechanisms for clinical T-DM1 resistance have been identified, less is known about innate or acquired resistance to DS-8201a. We established two XenoSTART Patient-Derived Xenograft (XPDX) models from tissue samples collected two years apart from a patient with ER+/HER2+ breast cancer before and after HER2 directed therapies. These models designated ST4565 and ST4565C were developed and characterized for receptor expression, genomic alterations, and in vivo drug sensitivities toward multiple chemotherapies and targeted agents, including T-DM1 and DS-8201a. Methods: Models ST4565 and ST4565C were established from breast samples collected from a Caucasian female with ER+/HER2+ metastatic breast cancer; ST4565 was collected at age 35 prior to therapy and ST4564C at age 37 following several treatment regimens including 5-FU/doxorubicin/cyclophosphamide, docetaxel/trastuzumab/pertuzumab, and T-DM1/anastrozole. Both were grown subcutaneously in female athymic nude mice supplemented with exogenous estradiol. The resulting models were passaged, and receptor expression confirmed immunohistochemically; genomic analysis, including WES and RNAseq, was performed to further characterize the models. For in vivo studies, both models were evaluated using several chemotherapy and targeted agents alone and in combination including: trastuzumab, pertuzumab, T-DM1, DS-8201a, neratinib, tucatinib, fulvestrant, alpelisib, sacituzumab, and irinotecan. In vivo study 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. Results: ST4565 and ST4565C retained comparable receptor expression (ER=2+/HER2=3+) over tested passages with similar histology compared to archival clinical samples. DNA/RNA sequencing identified several conserved variants including loss of CDKN2A/B and MTAP and a CNV=14 for CCND1. In vivo, ST4565 and ST4565C were found resistant to T-DM1 at 3 mg/kg weekly with a T/C of 100% in both models. However, DS-8201a treatment at 3 mg/kg weekly resulted in partial tumor regressions in ST4565 (T/C=-51%) while ST4565C was found resistant (T/C=49%). Both models were found resistant to all tested chemotherapies and all other targeted therapies but reported similar sensitivity to fulvestrant (T/C=~40%). Conclusion: We established two XPDX models representing T-DM1-resistant, ER+/HER2+ breast cancer from breast samples collected two years apart from the same patient that were found differentially responsive to DS-8201a. These models can be utilized as a valuable tool in better understanding innate resistance to T-DM1 and acquired resistance to DS-8201a. Citation Format: George Plasko, Johnnie Flores, Alyssa Simonson, Peter Forofontov, Ashwin Varma, Amy Lang, Gladys Rodriguez, Kyriakos P. Papadopoulos, Drew Rasco, Amita Patnaik, Bruce Conway, Joe Johnston, Michael Wick. Establishment and characterization of two T-DM1-resistant, ER+/HER2+ breast XPDX models developed sequentially from the same patient with differential in vivo sensitivity to trastuzumab deruxtecan (DS-8201a) [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P1-13-03.

Background: Several CDK4/6 inhibitors have recently been approved in combination with letrozole or fulvestrant in hormone receptor-positive breast cancer. Although this combination therapy has been found effective in some patients, resistance often develops. To aid in developing new therapies for CDK4/6i-resistant breast cancer and better understand potential resistance mechanisms, we established two XenoSTART Patient-Derived Xenograft (XPDX) models representing ER+/HER2- breast cancer from tissue samples collected seventeen months apart from the same patient before and after palbociclib therapy. These models designated ST4887 and ST4887B were developed and characterized for receptor expression, genomic alterations, and in vivo drug sensitivities toward multiple chemotherapies and targeted agents, including CDK4/6i and fulvestrant. Methods: Models ST4887 and ST4887B were established from metastatic samples collected from a Caucasian female with ER+/HER2- metastatic breast cancer; ST4887 was collected at age 38 from a femur mass biopsy following several treatment regimens including paclitaxel/doxorubicin/cyclophosphamide, radiation and tamoxifen. ST4887B was collected at age 39 from a liver biopsy following treatment with palbociclib/letrozole then palbociclib/fulvestrant, and finally ixabepilone/capecitabine. Both were grown subcutaneously in female athymic nude mice supplemented with exogenous estradiol. The resulting models were passaged, and receptor expression confirmed immunohistochemically; genomic analysis, including WES and RNAseq, was performed to further characterize models. For in vivo studies, both models were evaluated using several chemotherapy and targeted agents alone and in combination including cisplatin, docetaxel, CDK4/6i, fulvestrant, letrozole, olaparib, niraparib, and sacituzumab. In vivo study 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. Results: ST4887 and ST4887B retained comparable receptor expression (ER=3+/HER2=1+) over tested passages with similar histology compared to archival clinical samples. DNA/RNA sequencing identified several conserved variants including a somatic BRCA2 truncation (BRCA2Y2660*); transcriptomic analysis revealed upregulation of several related genes but no notable fusions. In vivo, both models were insensitive to cisplatin or docetaxel, however ST4887 but not ST4887B was sensitive to fulvestrant or CDK4/6i therapies, although abemaciclib demonstrated some activity toward ST4887B. PARP inhibitors were active toward ST4887 and to a lesser extent ST4887B, while sacituzumab did not have a significant effect on either model. Conclusion: We established and characterized two XPDX models from the same patient before and after acquired resistance to the CDK4/6i palbociclib. Both models were found to retain receptor status and drug sensitivities similar to the patient at the time of sample collection. These models can be utilized as a valuable tool in better understanding acquired resistance to palbociclib. Citation Format: Ashwin Varma, Johnnie Flores, Alyssa Simonson, Anna Stackpole, Kyriakos P. Papadopoulos, Amita Patnaik, Drew Rasco, Muralidhar Beeram, Marisa Sandera, Michael Wick. Establishment and characterization of two ER+/HER2- XPDX models developed sequentially before and after acquired resistance to the CDK4/6 inhibitor palbociclib from a patient with metastatic breast cancer [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P1-13-22.

Background: Mechanisms of resistance to CDK4/6 inhibitors (CDK4/6i) have been well studied and several alterations identified including RB loss and altered expression of related genes including CCNE1, E2F4 and CDK6. However, whether duration of clinical treatment might elicit specific mechanism(s) of CDK4/6i resistance is unclear. To better understand if duration of clinical treatment correlates with unique resistance mechanisms, we established, characterized, and compared a panel of ER+ breast XPDX models from patients who benefitted then progressed on a CDK4/6i. Patients were separated into two groups by time to progression (TTP): those who responded up to twelve months (RES12) and patients with clinical response greater than one year (RES13+). Methods: Seventeen breast cancer XPDX models were analyzed, including eight previously described (7xRES12; 1xRES13+: SABCS2021: T Hernandez et al). Nine new models were established from seven patients: six from fluid samples with three designated as ductal (ST3105B, ST3105C, STM001B) and three lobular carcinoma (STM182, STM229, STM229B); two from lymph node biopsies (ST5676, STM127) and one from a liver core biopsy (ST4887B), all reported as ductal carcinoma. STM182 was classified as RES12 and the remaining eight as RES13+. These models were passaged and challenged with CDK4/6i to confirm resistance. Receptor expression was determined by IHC and genomic analyses including WES and RNAseq, were performed to identify mechanisms of resistance. For in vivo studies, CDK4/6i were dosed PO once daily at 50 mg/kg; endpoints included tumor volume (TV) and time from treatment initiation (TTI) 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 Results: Clinical TTP for RES12 (n=8) was four to twelve months and RES13+ (n=9) from thirteen to forty-two months. All models retained ER expression in evaluated passages with similar histology compared with archival clinical samples. Sequencing identified several variants including RB1 truncations or deletions and increased gene expression in CCND1, CCNE1 and the PIK3CA/AKT pathway. Interestingly, 5/8 RES12 models reported ESR1 mutations or fusions versus 1/9 RES13+ and PIK3CA mutations were reported in 1/8 RES12 versus 5/9 RES13+. Several RES13+ models also reported variants and increased amplification in the RICTOR/TORC2 pathway versus RES12. Conclusion: We have established, characterized, and compared a panel of seventeen breast XPDX models from fourteen female patients representing early or late acquired resistance to CDK4/6i therapy and identified potential differences in each set. These models and resulting data are useful in developing novel therapies for CDK4/6i-resistant patients. Citation Format: Alyssa Simonson, Johnnie Flores, Morgan Lynch, Emily Carpenter, Justine Hruzek, Jim Lund, Natalia Baños Herraiz, Kyriakos Papadopoulos, Amy Vander Woude, Gladys Rodriguez, Sreenivasa Chandana, Thomas Gribbin, Nehal Lakhani, Tatiana Hernandez, Maria Jose de Miguel, Amy Lang, Michael J. Wick. Short or long-term treatment with CDK4/6 inhibitors in patients with ER+ breast cancer: characterization and comparative analysis of resistance in seventeen XPDX models. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3869.

Background: Preclinical models of prostate cancer are challenging to develop and maintain, especially those that grow in a castrated setting while maintaining receptor and antigen expression. To this end, we have established and characterized a panel of XenoSTART patient-derived (XPDX) prostate models using both intact and castrated athymic nude mice. These models, designated ST1273, ST2347, ST4017, and ST4420, were characterized for receptor expression, genomic alterations, and in vivo drug sensitivity to relevant therapies. Methods: XPDX models representing prostate cancer were established from primary (ST1273) or metastatic (ST2347, ST4017, ST4420) biopsy samples implanted into intact athymic nude mice supplemented with exogenous testosterone. Resulting models were passaged and further developed in both intact and castrated athymic nudes until growth stabilization. Resulting models were characterized using genomic analysis, including WES and RNAseq, receptor expression, and in vivo drug sensitivity studies. Models found sensitive to enzalutamide were conditioned to resistance in vivo by chronic drug administration and resulting models (designated STxxxx/EZR) were characterized and compared with parent lines. For in vivo studies, activity of relevant treatments were benchmarked including enzalutamide administered once daily by oral gavage at 50 mg/kg and docetaxel administered by intravenous injection once weekly at 10 mg/kg. In vivo study 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. Results: Each model developed in a castrate or conditioned setting retained similar receptor expression to the parent model including positive AR (2+/3+) and PSMA (2+/3+) staining. Genomic characterization identified a PIK3CA mutation in ST1273 models (PIK3CAE545A), AR mutations in ST2347 (ART878A) and ST4017 (ARH875Y), and a TMPRSS2:ERG fusion in ST4420. In vivo, the ST1273 parent model was found sensitive to enzalutamide (%T/C=11%) but insensitive in castrated mice (%T/C=76%) and the ST1273/EZR model was resistant to enzalutamide in intact (%T/C=56%) or castrated (%T/C=100%) mice. The ST2347 parent model was also found sensitive to enzalutamide (%T/C=19%) but insensitive in castrated mice (%T/C=50%). ST4017 and ST4420 studies in intact and castrated mice are currently underway. All models were sensitive to docetaxel. Conclusion: We have established and characterized a panel of prostate XPDX models using both intact and castrated athymic nude mice and conditioned resistance to enzalutamide by chronic drug administration. These models can be utilized as a valuable tool in better understanding castrate-resistant prostate cancer and in developing novel therapies for enzalutamide-resistant patients. Citation Format: Johnnie Flores, Alyssa Simonson, Dustin Kneifel, Alejandra Diaz, Morgan Harris, Kyriakos Papadopoulos, Amita Patnaik, Drew Rasco, Scott Ulmer, Michael J. Wick. Establishment and characterization of a panel of castrate-resistant prostate cancer XPDX models with differential enzalutamide response [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3109.

Background: First-line treatment for some cancer types includes platinum-based therapy. While the rate of initial response to treatment is high, most patients develop platinum-resistant disease. Current salvage therapy provides benefit to some patients, demonstrating the need for additional effective therapies. To assist in identifying new therapies, we have established XenoSTART patient-derived xenograft (XPDX) models representing breast, ovary, uterine, and lung cancers. Each model was characterized by DNA/RNA analysis, sensitivity to platinum treatment, and annotated with donor patient treatment status at the time of sample collection. Methods: 210 XPDX models were evaluated in this screen including 66 breast, 56 ovary, 36 uterine, and 52 lung xenografts. WES and RNAseq were performed on each model and patient treatment history and outcome annotated. For in vivo studies, models were implanted into female nudes and administered (IP; q7dx3) 3 m/k cisplatin or 60 m/k carboplatin. Study 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. Results: Study results are summarized below in Table 1: Sequencing identified several variants in resistant, chemo-naïve models including point mutations in RAS/RAF/MET/PIK3CA genes and others, while sensitive models lacked common driver mutations. Conclusion: We have characterized 210 XPDX models. Uterine models were most often resistant to platinum, while almost 1/2 of ovary and 1/3 of breast models were sensitive, regardless of treatment status. Only 15% of lung models were sensitive to platinum; cancer driver variants were found in several models. Overall, we report differential platinum sensitivity in a panel of diverse models useful in understanding mechanisms of resistance and for development of effective therapies in platinum-resistant cancers. Table 1. Type Breast Ovary Uterine Lung Total 66 56 36 52 # Sensitive 25 25 9 8 % 38% 45% 25% 15% Naïve 11 16 8 5 % 44% 64% 89% 63% P-1st 6 6 1 2 % 24% 24% 11% 25% P-2nd+ 6 3 0 1 % 24% 12% 0% 13% # Insensitive 41 31 27 44 % 62% 55% 75% 85% Naïve 11 9 20 21 % 27% 29% 74% 48% P-1st 14 10 5 13 % 34% 32% 19% 30% P-2nd+ 16 12 2 10 % 39% 39% 7% 23% Naïve=No Prior Treatment; P-1st=Post 1st Line Therapy; P-2nd=Post 2nd+ Line Therapy Citation Format: Alyssa Simonson, Johnnie Flores, Lizette Firova, Christian Hernandez, Morgan Harris, Kyriakos Papadopoulos, Drew Rasco, Amita Patnaik, Allan White, Lon Smith, Ronald Drengler, Amy Lang, Murali Beeram, Michael J. Wick. Correlation of platinum sensitivity with donor patient treatment status in a panel of breast, ovary, uterine, and lung XPDX models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1092.

Background: Neuregulin-1 (NRG1) rearrangements drive cancers by binding to ERBB3/ERBB2 heterodimers and activating downstream signaling. NRG1 fusion proteins have recently been identified in several cancer types including lung and ovary. To better understand the role of NRG1 fusions in cancer we established three XenoSTART patient-derived xenograft (XPDX) models driven by NRG1 rearrangements including two CD74-NRG1 lung models designated ST2891 and ST3204 and one SARAF-NRG1 ovary model designated ST2476. These models were established in athymic nude mice and characterized for receptor expression, genomic alterations, and in vivo drug sensitivity. Methods: ST2891 was established from an excision biopsy collected from a 68-year-old male with NSCLC following treatment with carboplatin/paclitaxel and carboplatin/pemetrexed. ST3204 was established from a lymph node biopsy collected from a 63-year-old female with NSCLC following treatment with cisplatin/pemetrexed. ST2476 was established from primary tissue collected from a chemo naïve 61-year-old female with ovarian cancer. The resulting models were passaged and characterized using immunohistochemical and genomic analysis including WES and RNAseq. In vivo studies were performed testing various chemotherapy and targeted agents; study 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 Results: All models reported ERBB2 and ERBB3 staining in evaluated passages with similar histology compared with archival clinical samples. WES analysis of the three models did not identify any variants in EGFR/ERBB2-4/RAS/RAF genes. RNA sequencing revealed CD74-NRG1 or SARAF-NRG1 gene fusions independently confirmed using molecular studies. In vivo, both lung models were resistant to weekly cisplatin or T-DM1 and daily palbociclib (50 mg/kg) or daily afatinib up to 10 mg/kg. ST3204 was found sensitive to weekly pertuzumab (1 mg/kg). Both lung models were resistant to trastuzumab (1 mg/kg). The ST2476 ovary XPDX was found resistant to endocrine therapies and reported some sensitivity to weekly cisplatin. Conclusion: We have established and characterized a panel of three XPDX models driven by NRG1 rearrangements including two CD74-NRG1 lung models and one SARAF-NRG1 ovary model. These models are valuable tools for further developing therapies targeting NRG1-driven cancers. Citation Format: Alyssa Simonson, Johnnie Flores, Crystal Moreno, Justine Hruzek, Kyriakos Papadopoulos, Drew Rasco, Amita Patnaik, Ronald Drengler, Allan White, Jun Ma, Michael J. Wick. Establishment and characterization of neuregulin-1 (NRG1) fusion driven XPDX models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 353.

Background: We previously reported (RAS AACR 2022) a preclinical screen testing KRASG12C-inhibitors (KRASG12C-i) in 13 colorectal and 13 lung KRASG12C XPDX models; all tested models were from patients naïve to KRASG12C-i. To better understand mechanisms of resistance in this agent class, we established three lung XPDX models designated ST5185B, ST5431B, and ST5489 from patients before therapy with a KRASG12C-i. We also established one colorectal XPDX, designated ST4859, post therapy. Once established these models were molecularly characterized, in vivo sensitivity to sotorasib and adagrasib determined, and results compared with clinical response. Ex vivo cultures and 3D-XPDX࣪ studies were also performed evaluating both compounds comparing drug sensitivity. Methods: Lung: ST5185B was established from a pretreated 68-year-old male; ST5431B was established from a chemo-naïve 67-year-old male; additional samples were collected during and after progression on KRASG12C-i and are currently in development. ST5489 was established from a pretreated 64-year-old female. CRC: ST4859 was established from a 57-year-old female pretreated with a KRASG12C-i; all patients received at least one cycle of KRASG12C-i therapy. Models were WES and RNA sequenced and sotorasib and adagrasib sensitivity determined. For each study, agents were dosed PO once daily at 100 mg/kg; study endpoints included tumor volume and time from treatment initiation with %T/C values reported at study completion; a %T/C of ≤ 20% was considered sensitive. Results: In all models sequencing confirmed KRASG12C and other variants previously identified with clinical studies. In vivo studies identified ST5185B as resistant to sotorasib (%T/C=70%) and adagrasib (%T/C=86%); this donor patient progressed after six weeks on KRASG12C-i therapy. ST5431B reported sensitivity to both therapies; this donor patient had a partial response (50% decrease) for six months prior to progression. ST5489 reported sensitivity to sotorasib (%T/C=14%) and adagrasib (%T/C=8%); this donor patient interrupted dosing due to KRASG12C-i intolerance and discontinued therapy, with disease progression after five weeks. ST4859 reported resistance to both drugs; this donor patient had stable disease on KRASG12C-i therapy for approximately seven months prior to progression. Ex vivo cultures and 3D-XPDX࣪ studies from all models reported sensitivity to both agents similar to in vivo studies. Summary: We developed, evaluated, and correlated in vivo sensitivity of sotorasib and adagrasib in a panel of four XPDX models established from patients receiving KRASG12C-i therapy. Studies are underway to generate drug-resistant clones of sensitive models and to elucidate mechanisms of drug resistance in ST4859. Citation Format: Johnnie Flores, Alyssa Simonson, Peter Forofontov, Anna Stackpole, Drew Rasco, Amita Patnaik, Kyriakos Papadopoulos, Teresa DesRochers, Natalie Williams, Ronald Drengler, Lon Smith, Michael J. Wick. Correlation of drug sensitivity to clinical response in XPDX models established from patients treated with KRAS-G12C-inhibitor therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 412.

Background: 15-20% of metastatic breast cancers are characterized by overexpression or amplification of human epidermal growth factor receptor 2 (HER2). First-line treatment for HER2-positive metastatic breast cancer includes the anti-HER2 monoclonal antibodies trastuzumab and pertuzumab in combination with chemotherapy. Second-line therapy includes T-DM1, an antibody-drug conjugate (ADC) consisting of trastuzumab conjugated to maytansinoid (DM1). Until recently, patients who progress on T-DM1 were left with few treatment options. Trastuzumab deruxtecan (DS-8201a) is an antibody-drug conjugate (ADC) consisting of an anti-HER2 (human epidermal growth factor receptor 2) antibody linked to a novel topoisomerase I inhibitor payload using a cleavable tetrapeptide-based linker and was recently approved for unresectable or T-DM1-refractory HER2+ breast cancer. In addition, DS-8201a is currently being evaluated as a treatment in breast cancers with low and medium expression of HER2. To better understand the potential for DS-8201a as a treatment in these cancer types, we evaluated this therapy in a panel of ER+/- XPDX models with differential HER2 expression. In addition, we compared DS-8201a activity to T-DM1 and benchmarked efficacy of the topoisomerase I inhibitor irinotecan in all tested models. Methods: One hundred (50ER+/50ER-) breast XPDX models were evaluated in this study. Models were established and characterized for estrogen receptor and HER2 expression by IHC and profiled using WES and RNAseq. For in vivo studies, DS-8201a and T-DM1 were administered IV at 3 mg/kg and irinotecan IP at 100 mg/kg on a weeklyx3 schedule. In vivo study 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. Results: Models were grouped by ER positivity (+/-) and model HER2 score and designated as null (0), low (1+), medium (2+) or high (3+); models established from clinically HER2+ patients were also noted. In the ER+ group, HER2 low and medium expressing models accounted for over 80% of the total, while 15% were categorized as high and less than 5% as null. All models with high and 30% with medium HER2 staining were from clinically HER2+ patients, no low or null models were established from breast cancer assigned HER2+ clinically. In the ER- group, HER2 low and medium expressing models accounted for 75% of the total, while 12% were categorized as high and 13% as null. All models with high and 13% with medium HER2 staining were from clinically HER2+ patients; 5% of low or null models were established from breast cancer assigned HER2+ clinically. In vivo, 65% of the ER+ group reported sensitivity to DS-8201a versus 25% to T-DM1; 10% of DS-8201a, and 40% of T-DM1 sensitive models were high expression models and the remaining low and medium with no null models reporting sensitivity to either agent. 35% of tested ER+ models were sensitive to both HER2 therapies and 50% of models sensitive to DS-8201a were also sensitive to irinotecan. In the ER- group, 70% reported sensitivity to DS-8201a versus 25% to T-DM1; 20% of DS-8201a and 25% of T-DM1 sensitive models were high expression models and the remaining low and medium with one null model (ST069) sensitive to both agents. 25% of tested ER- models were sensitive to both HER2 therapies and 40% of models sensitive to DS-8201a were also sensitive to irinotecan. Conclusion: We have compared activity of DS-8201a, T-DM1 and irinotecan in a panel of ER+/- XPDX models and correlated activity based on HER2 expression. This data and panel can be utilized as a valuable tool in better understanding the potential for DS-8201a and other HER2-targeting therapies as a treatment in cancers driven by HER2 expression. Citation Format: Alyssa Simonson, Peter Forofontov, Johnnie R Flores, Kimberly Hernandez, April Cabang, Amy Lang, Gladys Rodriguez, Kyriakos Papadopoulos, Murali Beeram, Arthur Rosenthal, Brittany DeBerry, Lon Smith, Ronald Drengler, Amita Patnaik, Drew Rasco, Luis Rodriguez, Steven Abbate, Scott Ulmer, Michael Wick. Correlation of HER2 receptor expression and in vivo activity of the HER2-targeting therapies trastuzumab deruxtecan (DS-8201a) and T-DM1 activity in a panel of breast XPDX models [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P5-01-04.

Background: Trastuzumab deruxtecan (DS-8201a) is an antibody-drug conjugate (ADC) consisting of an anti-HER2 (human epidermal growth factor receptor 2) antibody linked to a topoisomerase I inhibitor payload using a cleavable tetrapeptide-based linker and was recently approved for unresectable or T-DM1-refractory HER2+ breast cancer. While some mechanisms for clinical T-DM1 resistance have been identified, less is known about acquired or innate resistance to DS-8201a. We established two XPDX models of ER+/HER2+ breast cancer from tissue and fluid samples collected simultaneously from the same patient. These models designated ST4480B and ST4480C were developed and characterized for receptor expression, genomic alterations, and in vivo drug sensitivities toward multiple chemotherapies and targeted agents including DS-8201a and T-DM1. Methods: ST4480B and ST4480C were established from a 70-year-old Caucasian female with ER+/HER2+ metastatic breast cancer pretreated with chemotherapy and targeted agents including T-DM1 for nine months followed by capecitabine/trastuzumab/tucatinib combination for one year prior to sample collections. ST4480B was established from a lymph node biopsy and ST4480C from a fluid sample collected the same day; both were grown subcutaneously in female athymic nude mice supplemented with estradiol. The resulting models were passaged and receptor expression confirmed immunohistochemically; genomic analysis, including WES and RNAseq, was performed to further characterize models. For in vivo studies, both models were evaluated with several chemotherapy and targeted agents alone and in combination including: trastuzumab, pertuzumab, T-DM1, DS-8201a, neratinib, tucatinib, alpelisib, everolimus, and irinotecan. In vivo study 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=— Citation Format: Johnnie R Flores, Anna Stackpole, Abimael Garza, Alexandra Ulmer, Alyssa Simonson, Kyriakos Papadopoulos, April Cabang, Jun Ma, Amita Patnaik, Drew Rasco, Amy Lang, Gladys Rodriguez, Murali Beeram, Michael J Wick. Establishment and characterization of two simultaneously developed T-DM1-resistant, ER+/HER2+ XPDX models from the same patient with differential in vivo sensitivity to trastuzumab deruxtecan (DS-8201a) [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P5-01-09.

Background: Fulvestrant is a selective estrogen receptor modulator (SERM) approved as a single agent for estrogen receptor positive, HER2 negative breast cancer patients at various stages of disease and in combination with CDK4/6 inhibitors following endocrine therapy failure. Although this agent requires intramuscular injection, it has demonstrated superior activity and fewer side effects versus some oral endocrine treatments however, resistance to fulvestrant often develops. To better understand fulvestrant resistance and its utility in patients who have failed other endocrine therapies, we evaluated the agent in a panel of ER+ breast models established from patients at various stages of disease representing endocrine-sensitive and -resistant disease. Methods: Sixty-five previously developed ER+ breast XPDX models were evaluated in this study. Models were grown subcutaneously in female athymic nude mice supplemented with estradiol in drinking water when necessary. All models were characterized at early and late passages for estrogen receptor expression by immunohistochemistry and profiled using WES and RNAseq. For in vivo studies, fulvestrant was administered by subcutaneous injection at 2.5 or 5 mg per dose once weekly until study completion. In vivo study 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 Citation Format: April Cabang, Crystal Moreno, Johnnie R Flores, Jenna Boedeker, Alyssa Simonson, Jun Ma, Amy Lang, Gladys Rodriguez, Arthur Rosenthal, Kyriakos Papadopoulos, Amita Patnaik, Drew Rasco, Lon Smith, Murali Beeram, Ronald Drengler, Luis Rodriguez, Steven Abbate, Scott Ulmer, Michael J Wick. Nonclinical activity of fulvestrant in a panel of ER+ breast XPDX models representing clinically acquired and innate resistance to endocrine therapies [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P5-01-11.

Background: Several CDK 4/6 inhibitors have recently been approved in combination with letrozole or fulvestrant in hormone receptor-positive breast cancer. Although this combination therapy has been found effective in some patients, resistance often develops. To aid in developing new therapies for CDK4/6 inhibitor-resistant breast cancer and better understand potential resistance mechanisms, we established a panel of nine XPDX models from eight female patients with luminal A breast cancer at time of progression following acquired resistance to CDK4/6 inhibitor therapy. These models, designated ST940C, ST2056, ST3164B, ST3164B/PBR, ST3932, ST4316B, ST4378, STF160, and STM001B were developed in athymic nude mice and characterized for receptor expression, genomic alterations, and in vivo drug sensitivity. Methods: STF160 was established from a primary biopsy and ST3932 from a metastatic soft tissue lesion; the remaining models were established from malignant fluid samples collected at various stages of treatment post CDK4/6 inhibitor response and progression. The resulting models were passaged and challenged with CDK4/6 inhibitors to confirm resistance and fulvestrant to assess sensitivity. Receptor expression was determined immunohistochemically. Genomic analysis, including WES and RNAseq, were performed to characterize models and identify mechanisms of resistance. For in vivo studies, palbociclib and abemaciclib were dosed by oral administration once daily at 50 mg/kg and fulvestrant by subcutaneous administration once weekly at 2.5 mg. In vivo study 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/C20) to palbociclib or abemaciclib and single agent fulvestrant. ST940C and ST2056 were sensitive (%T/C≤20) to tested CDK4/6 inhibitors but resistant to fulvestrant. Conclusion: We have established and characterized a panel of nine XPDX models from eight female patients with luminal A breast cancer at time of progression following acquired resistance to CDK4/6 inhibitor therapy, seven of which were found resistant to single agent palbociclib and abemaciclib and all nine to fulvestrant. This panel can be utilized as a valuable tool in better understanding CDK4/6 inhibitor resistance and in developing novel therapies for CDK4/6 inhibitor-resistant patients. Citation Format: Tatiana Hernandez, Dustin Kneifel, Alyssa Simonson, Johnnie R Flores, Sarah Quick, April Cabang, Alexandra Ulmer, Kyriakos Papadopoulos, Amy Lang, Gladys Rodriguez, Murali Beeram, Drew Rasco, Amita Patnaik, Scott Ulmer, Michael J Wick. Establishment and characterization of luminal A breast XPDX models from patients with acquired resistance to CDK 4/6 inhibitors [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P5-01-06.