Yusra Shao
Has grant
Assistant Professor
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Recent Grants
Grant: Close
Melanoma Breslow Thickness Pre, During and Post COVID-Lockdown
Open Date:
Close Date:
Grant: Close
Nuclear Export Protein XPO1 Inhibition in BAP1 Mutated and Wildtype Uveal Melanoma
Open Date:
Close Date:
Articles (11)
Background: Dysregulation of hypoxia-inducible factor 2-alpha (HIF-2α) in clear-cell renal cell carcinoma (ccRCC) contributes to increased angiogenesis, proliferation, and cancer cell survival. Casdatifan (Cas) is an orally bioavailable small-molecule HIF-2α inhibitor that has shown promising efficacy in metastatic ccRCC in the ARC-20 phase 1 trial (NCT05536141). Erythropoietin (EPO) is a known HIF-2α transcription target, and serum EPO (sEPO) levels decrease after treatment with HIF-2α inhibitors. However, the relationship between EPO expression in ccRCC patients, HIF-2α inhibitor-induced changes and clinical benefit has not been studied. Here we investigated baseline and longitudinal sEPO levels and tumor EPO expression in ARC-20 as potential biomarkers for systemic HIF-2α inhibition and explored their association with clinical benefit of Cas. Methods: The biomarker-evaluable population (BEP) was comprised of 127 patients evaluable for overall response (as of 3/14/25) and enrolled in the 50-200mg Cas monotherapy dose escalation and expansion cohorts. sEPO was measured by immunoassay; tumor EPO mRNA was quantified by RNAseq. Results: Baseline sEPO levels from BEP were significantly higher than healthy volunteers (P<0.0001) and above the normal sEPO range (>18.5mIU/ml) in 49% (62/127) of BEP. Following Cas treatment, sEPO levels decreased as early as 4hr after the first dose and reached maximal suppression between day 8 and day 22 on treatment. Deep sEPO reductions were observed at all Cas dose levels and sustained throughout the course of treatment. Some patients experienced a slight sEPO rebound to ∼ -60% of baseline levels at 12-week post-treatment, following which sEPO suppression was maintained for the duration of treatment. The maximum sEPO decreases for the 100mg QD (n=28) cohort ranged from -99% to -47%. In 2 pairs of matched pre-/post-treatment biopsies, expression of tumor EPO and HIF-2α transcription target genes, SERPINE1 and PTHLH, were downregulated 2 to 61-fold after Cas treatment. Patients with an objective response had significantly deeper maximum sEPO reduction on treatment compared to patients with progressive disease (PD) as their best overall response (P<0.001). Consistently, patients with a deeper (by median, -85%) maximum sEPO reduction had a higher overall response rate (ORR) (37.5%) and a lower primary PD rate (7.8%) than those with less deep sEPO reductions (ORR: 17.5%; PD: 23.8%), indicating a strong association between depth of sEPO reduction and clinical benefit. Furthermore, among patients with elevated (>18.5mIU/ml) baseline sEPO levels, only 1/14 (7%) with high tumor EPO expression experienced primary PD on Cas compared to 4/12 (33%) cases with low tumor EPO expression. Conclusions: sEPO may provide a window into systemic HIF-2α inhibition in ccRCC patients. Cas monotherapy resulted in deep and sustained reductions of sEPO, as well as a significant downregulation of tumor EPO and other HIF-2α target genes. Depth of sEPO reduction and tumor EPO expression were strongly associated with clinical benefit for Cas. Citation Format: Yinghui Guan, Ben Weeder, Amita Patnaik, Moshe Ornstein, Benjamin Garmezy, Alexandra Drakaki, Brian Rini, Jamie Merchan, Sun Young Rha, Ralph Hauke, Rohit Kumar, Pedro Barata, Craig Gedye, Che-Kai Tsao, Jae Lyun. Lee, Yusra Shao, Se Hoon. Park, Marc Matrana, Clara Hwang, Hunter Cole, Chris Negro, Paul Foster, Mohammad Ghasemi, Jianfen Chen, Toni Choueiri, Sean Cho, Omar Kabbarah. Erythropoietin (EPO) is a Pharmacodynamic Biomarker for Systemic HIF-2α Inhibition that Correlates with the Clinical Activity of Casdatifan [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2025 Oct 22-26; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2025;24(10 Suppl):Abstract nr A023.
Year:
2025
Year:
2025
Year:
2025
Year:
2025
e21598 Background: While large language models (LLMs) like GPT-4 (OpenAI) and Claude Opus (Anthropic) perform well on general medical exams, limited data exist on their application in oncology, particularly in multiple-choice question (MCQ) scenarios. This study benchmarks the accuracy and assesses the clinical reasoning of LLMs on cutaneous oncology MCQs from the American Society of Clinical Oncology (ASCO) question bank. Methods: Skin oncology MCQs from the ASCO question bank were tested using GPT-4 (OpenAI) and Claude Opus (Anthropic). Questions were entered into OpenAI and Anthropic API playgrounds without additional prompting, with identical settings for both models (Temperature = 0, Token = Max). Each question was tested three times per model to assess response consistency and precision. Initial responses were collected, and chain-of-thought (COT) prompting was applied to encourage stepwise reasoning mimicking an oncologist’s thought process. The percentage of revised answers after COT prompting was analyzed. Accuracy before and after COT prompting was compared. Incorrect responses were reviewed by board-certified oncologists specializing in skin cancer at National Cancer Institute (NCI)-designated centers. Each oncologist scored the responses on clarity, clinical reasoning, bias, and confidence in the models’ utility in clinical practice. Qualitative feedback was collected via free-text comments and analyzed descriptively. Results: Of 66 skin oncology MCQs, Claude correctly answered 55 questions (83% accuracy, 95% CI: 72%–91%), while ChatGPT correctly answered 53 (80% accuracy, 95% CI: 68%–89%). Both models performed comparably, with overlapping confidence intervals. A chi-square test showed no significant difference between their performances (p = 0.65). After COT prompting, both tools remained at their respective accuracy level. Thematic analysis of incorrect answers revealed that LLM responses frequently deviated from clinical guidelines and failed to reflect evidence-based oncology practice, often neglecting key trial data like DREAMSeq and frontline therapeutic strategies. The most frequent errors were Treatment Planning Errors, where recommendations misaligned with NCCN guidelines, and Interpretation Errors, involving misinterpretation of clinical trials or patient data. Anchoring bias emerged as a recurring issue, with models fixating on details (e.g., BRAF mutation status) at the expense of broader reasoning. Conclusions: This study demonstrates that while large language models like GPT-4 and Claude Opus show promise in the field of oncology, they still face significant challenges in clinical reasoning and adherence to evidence-based practices. Continued refinement and training on specialized medical data are essential before these models can be reliably integrated into clinical workflows.
Year:
2025
Year:
2025
Background: Dysregulated nuclear protein transport, commonly observed in cancer, results in the mislocalization-mediated inactivation of many proteins. We showed that overexpression of exportin 1 (XPO1) is linked to higher grade and Gleason score in metastatic castration-resistant prostate cancer (mCRPC). Using network topology computational approaches, XPO1 and PARP1 were identified as synthetic lethal partners and we have demonstrated synergy between XPO1 and PARP inhibitors in mCRPC cell lines. This study evaluates the combination’s efficacy in cell line-derived (CDX) and patient-derived xenograft (PDX) models, investigating the mechanisms of synergy. Methods: For the CDx and PDx models, 22rv1 cells and CTG-3581 tissue (Champions Oncology) were grown subcutaneously in ICR-SCID and CEIA/NOG male mice, respectively. XPO1 inhibitor was dosed orally at 10-15 mg/kg twice a week and PARP inhibitor was dosed orally at 50 mg/kg daily. For in vitro mechanistic study, 22rv1 cells were subjected to RNAseq, proteomic analysis, and Digital Spatial Profiling (DSP; 10x Genomics Visium platform) after treatment. Data analysis was performed using iPathwayGuide (advaitabio.com) and cLoupe browser. Results: The XPO1-PARP inhibitors combination showed enhanced anti-cancer efficacy in both CDx and PDx models, with no significant weight loss or organ toxicity such as liver, spleen, and kidney. Survival benefits were evident in the combination group. Immunohistochemistry revealed apoptotic cell death, shown by cleaved caspase 3 and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining. Mechanistic studies with 22rv1 cells showed downregulation of DNA replication genes (e.g., MCM6 and CDC6) and reduced enrichment scores for DNA replication pathways in gene set enrichment analysis. The proteomic analysis highlighted decreased levels of DNA replication modulators HMGB2 and DNAJC9. Besides the suppression of various DNA damage response genes (e.g. BRCA2 and RAD51), XPO1 inhibition also reduced the PARylation of mCRPC cells. DSP analysis of 22rv1 CDx tissue revealed a reduction of unbiased cellular clusters and increased differentially expressed genes in the combination treatment. The DNA replication pathway was among the top 5 pathways modulated by XPO1i treatment in DSP analysis. Conclusions: Taken together, this study revealed the therapeutic potential of XPO1-PARP inhibitors combination via targeting of the DNA damage response pathway in mCRPC. Citation Format: Md. Hafiz Uddin, Aaban A. Azmi, Laiba M. Monir, Amro Aboukameel, Husain Y. Khan, Sahar F. Bannoura, Frank Cackowski, Yusra Shao, Vinod Shidham, Sunil Jaiman, Rafic Beydoun, Gregory Dyson, Seongho Kim, Yang Shi, Vy Ong, Julie Boerner, Ramzi M. Mohammad, Elisabeth I. Heath. XPO1 inhibition synergizes with PARP inhibitor in metastatic castration resistant prostate cancer via the regulation of DNA damage response [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5716.
Year:
2025
Responses to immune checkpoint inhibitors (ICIs) differ significantly between uveal melanoma (UM) and cutaneous melanoma (CM) patients. We investigated the immune profile of metastatic UM(mUM) patients and identified markers that are predictive of improved survival. Metastatic liver samples from 189 UM patients and 48 CM patients were analyzed at genomic and transcriptional levels, and survival analysis was performed on a subgroup of 76 ICI-treated mUM patients. UM liver metastases seem to preserve the genomic and immune characteristics of primary UM (pUM), with a low prevalence of ICI markers and high mutation rates of GNA11, GNAQ, BAP1, and SF3B. Compared with mCM, UM liver metastasis showed lower infiltration of several immune cells, but a greater proportion of M2 macrophages. Compared with UM liver metastases, CM liver metastases showed higher expression of G2M checkpoint and EF2 target genes. Among the mUM and mCM samples, expression of G2M and E2F pathway genes was highest in tumors with high TMB values and T-cell inflamed scores. A longer median overall survival (OS) was observed in mUM patients with higher expression of LAG3, HLA class I, or HLA class II; which may represent a small proportion of immune hot tumors. Expression patterns of G2M checkpoint and E2F targets suggest a possible contribution to immunotherapy response.
Year:
2025
e15158 Background: Autosomal dominant (AD) pathogenic/ likely pathogenic (P/LP) variants in the VHL gene cause Von Hippel Lindau syndrome, historically studied in White populations. We aimed to characterize VHL variants and disease phenotypes in both Non-White (NW) patients, which include Asian and African-American (AA), and in Non-Hispanic White (NHW) patients. Methods: Patients undergoing germline genetic testing (GGT) at Invitae Corporation from November 2014 to March 2022, carrying a P/LP AD variant in VHL gene were identified. Demographic and clinical data were obtained from Test requisition forms. Fisher's exact test and Wilcoxon rank-sum tests were performed for categorical and continuous variables, respectively, to compare groups. Results: The final analysis included 363 patients (NW=27%). Hemangioblastoma was predominant in all groups (NHW=13%, NW=27%, p=0.004). Breast cancer exclusively occurred in NHW patients (7%, p=0.019). NHW and NW patients had a higher family history of breast cancer (NHW=7%, NW=1%, p=0.046) and hemangioblastoma (NHW=3%, NW=8%, p=0.037), respectively. Distribution of VHL variants between NW and NHW patients varied significantly (p<0.001, Indels p=0.009, substitution variants p<0.001). Deletion exon 2 was predominant in all groups (NHW=15%, NW=35%) P.Leu188Val was exclusive to NHW patients (26%)while p.Arg167Trp (NHW=6%, NW=11%) and p.Arg167Gln (NHW=2%, NW=10%) were more common in NW patients. Several unique variants in AA and Hispanic patients were seen. Conclusions: This study reveals differences in P/LP VHL variant distribution and disease phenotype between NHW and NW races. Ethnicity-specific characterization uncovers unique variants in AA and Hispanic patients. This dataset informs future research on VHL disease and treatment in diverse populations.
Year:
2024
e21574 Background: Access to routine cancer care during COVID-19 pandemic was greatly affected. The impact of this on melanoma care is unknown. Breslow thickness (BT) was seen to decrease and then increase during and post-COVID-19 lockdown in Italy. Here, we aim to study the impact of COVID-19 lockdown on melanoma care in the United States. Methods: Patients diagnosed with cutaneous melanoma (CM) pre (September 15th 2019-March 14th 2020), during (March 15th-June 14th 2020), and post-lockdown (June 15th-December 14th 2020) were retrospectively studied using Metropolitan Detroit Cancer Surveillance System (MDCSS) database. Patients without an identifiable primary lesion, or with ocular, mucosal, or acral melanoma were excluded. Demographic and clinical characteristics were reviewed. The primary endpoint was median BT during and post vs pre-lockdown. Fisher's exact and Kruskal-Wallis tests were performed for categorical and continuous variables, respectively, to compare groups. Interaction in subgroup analyses was assessed using likelihood ratio tests. Linear regression analyses were performed with Box-Cox transformed BT. Cox proportional hazards regression analyses were performed for overall survival. Covariates for multivariable analysis were selected using LASSO-based penalized regression models. Results: 802 patients were included (pre N = 347, during N = 87, post N = 368). Patients diagnosed with CM post-lockdown were younger (age < 50 years pre = 14%, during = 13%, post = 23%). More patients diagnosed pre-lockdown carried government-issued insurance (pre = 48%, during = 39%, post = 38%). No difference (diff) in median BT was seen between periods (median = 0.6mm for pre, during, and pos). Married status was associated with a lower BT vs unmarried (diff = -0.19, 95% CI -0.36 to -0.014, p = 0.034) and higher census tract poverty level was associated with a higher BT (5%-10% diff = 0.17, 95% CI 0.004 to 0.33, p = 0.045; 10%-100% diff = 0.27, 95% CI 0.093 to 0.45, p = 0.003) vs 0-5% poverty on univariable analysis. Male gender was associated with a lower BT on multivariable analysis (diff = -0.12, 95% CI -0.20 to -0.04, p = 0.005) and lower risk of death on univariable analysis (HR = 0.50, 95% CI 0.31 to 0.82, p = 0.006). Conclusions: We found no difference in median BT during and post- vs pre-lockdown. We also identified at-risk populations (female, unmarried, and living in greater poverty areas) who could benefit from proactive melanoma care. Time to surgery from initial diagnosis during and post vs pre-lockdown would be of interest to study further.
Year:
2024
Year:
2023
Collaborators (1)
Edward Esplin
Clinical Geneticist
Stanford University
Make your next move.
Prepare your application
Review drafts before you send them.