Zakharia, Yousef

136 Background: The ECOG performance status (PS) is a critical measure for evaluating functional capacity and determining clinical trial eligibility in patients with advanced prostate cancer. We aim to assess the capability of large language models (LLMs) to accurately infer ECOG PS from unstructured oncology notes. Methods: This retrospective study included patients with metastatic castration-resistant prostate cancer (mCRPC) receiving Lutetium-177–PSMA-617 ( 177 Lu) at the Mayo Clinic (2022-2023). Baseline ECOG PS scores were manually curated by a trained clinician from clinical notes recorded within four weeks prior to 177 Lu initiation. Structured zero-shot prompts were then used to prompt GPT-4 to infer ECOG PS scores from unstructured oncology notes. Total dataset was randomly sampled to prompt development (20%) and held-out test set (80%). Three experimental setups were employed to assess model performance: (i) utilizing unmasked clinical notes, (ii) clinical notes with only numerical ECOG PS scores masked while retaining narrative descriptive/assessment by oncologist of ECOG score, and (iii) clinical notes with both numerical scores and narrative ECOG descriptors masked. Final performance was assessed using evaluation metrics (precision, recall, F1 score) on the held-out test set. Results: A total of 240 patients (n: 50 prompt development; n: 190 test set) were included in the analysis. The median age was 70 (IQR: 66-76); a majority of patients were White (n: 229; 95%) and non-Hispanic (n: 233; 97%). The most prevalent ECOG PS score was 0 (n: 140; 58%), followed by 1 (n: 88; 37%), and 2 (n: 12; 5%). The performance was numerically higher when using unmasked notes (weighted average precision: 0.89; recall: 0.89; F1: 0.89), compared to using notes with masked scores only (weighted average precision: 0.76; recall: 0.79; F1: 0.75) and notes with both scores and narrative masked (weighted average precision: 0.57; recall: 0.53; F1: 0.48). Metrics by each ECOG PS score is outlined (Table). Conclusions: Large language models demonstrate significant potential in precisely extracting ECOG performance status from unstructured oncology notes, offering valuable support for treatment decision-making and clinical trial enrollment. However, model performance declines as reliance on implicit reporting within clinical notes increases. Future efforts focusing on prediction of ECOG using a series of notes over time will be the next step. ECOG Unmasked oncology notes ECOG Score masked only ECOG Score + ECOG narrative marked Precision Recall F1 Score Precision Recall F1 Score Precision Recall F1 Score 0 0.89 0.92 0.91 0.75 0.93 0.83 0.59 0.74 0.66 1 0.88 0.86 0.87 0.76 0.74 0.75 0.19 0.47 0.27 2 1.00 0.82 0.90 0.78 0.33 0.47 0.78 0.16 0.26

134 Background: Prostate cancer patients with visceral metastases at diagnosis are often treated using one-size-fits-all approach. For example, CHAARTED criteria treats any visceral metastasis as high volume. Therefore, we aimed to assess overall survival (OS) by site of visceral involvement in patients with metastatic PCa. Methods: Surveillance, Epidemiology, and End Results (SEER) database (2010-2021) was queried to obtain case listing data on metastatic PCa patients from 17 registries. The patients were categorized into three groups based on visceral metastasis at diagnosis: lung metastasis, liver metastasis, and others. Kaplan-Meier survival analysis was performed to estimate median OS with 95% confidence interval (CI). Log-rank test was conducted to assess if survival was significantly different among the groups. Cox proportional hazard regression analysis was performed to assess the magnitude of difference among the groups. P-value <0.05 established statistical significance. Results: This analysis included 21,007 metastatic PCa patients (lung: n=319; liver: n=123; brain: 107; bone: 20538). Metastatic PCa population with lung metastasis exhibited a median OS of 60 months (95% CI: 48-72) compared to 24 months (95% CI: 12-36) in population with liver metastases and to 36 months (95% CI: 36-36) in population with other metastases. The difference in OS among the groups was statistically significant (p < 0.001). Compared to population with lung metastasis, those with liver (HR: 1.91; 95% CI: 1.41-2.57) but not those with other metastases (1.16; 0.97-1.39) had significantly worse OS adjusting for Gleason score, PSA and race (Table). Conclusions: In metastatic PCa, patients with lung-only metastasis have significantly better overall survival and should be considered differently in prognostication to liver metastases. Consideration of site of visceral metastasis is critical for treatment intensification decisions. Limitations of this analysis include lack of accounting for confounding relationships at individual patient level. Variable Hazard Ratio (95% CI) Metastases (Lung) REFERENCE Metastases (Liver) 1.91 (1.41-2.57) Metastases (Other) 1.16 (0.97-1.39) Gleason score≥8 REFERENCE Gleason score <8 0.68 (0.64 -0.72) PSA - ng/dL 1.01 (1.00 – 1.01) Race (White) REFERENCE Race (Black) 0.99 (0.94 – 1.05) Race (Asian or Pacific Islander) 0.70 (0.64 – 0.76)

132 Background: Distant metastasis at the time of prostate cancer (PCa) diagnosis is associated with increased morbidity and mortality. This study aimed to assess the impact of sociodemographic characteristics on distant metastasis at the time of PCa diagnosis. Methods: The Surveillance Epidemiology and End Results (SEER) database (2000-2021), including PCa patients from 17 registries, was queried. Patients with late-stage diagnosis (LSD) were defined as those who were initially diagnosed with stage IV disease, indicating distant metastasis. Sociodemographic predictors included race, ethnicity, age, marital status, residential status, and median household income. Multicollinearity was assessed using variance inflation factors (VIF). Univariable logistic regression was performed to identify significant predictors of LSD, which were further analyzed in multivariable logistic regression using a maximum likelihood estimation approach. Odds ratios with 95% confidence intervals (CI) were computed with a p-value of <0.05 indicating a statistically significant association. Results: Of the 1,089,736 PCa patients diagnosed, 64,983 (6%) had an LSD. The majority were white (80%), non-Hispanic/Latino (HL) (91%), aged ≥65 years (60%), married (75%), residents in metropolitan areas (88%), and had an annual house income of ≥$70,000 (70%). Multivariable logistic regression identified Black (OR: 1.24; 95%CI: 1.22-1.27), Asian or Pacific Islander (1.35; 1.30-1.40), and American Indian/Alaska Native (1.70; 1.53-1.89) to have higher odds of LSD compared to White patients (Table). In addition, HL ethnicity compared to non-HL (OR: 1.45; 95% CI: 1.42-1.49), age ≥65 compared to <65 (1.68; 1.65-1.71), unmarried status compared to married (1.88; 1.85-1.91), residency in non-metropolitan areas compared to metropolitan (1.10; 1.07-1.12), and household income <$70,000 compared to ≥$70,000 (1.03; 1.01-1.05) had a significant association with LSD. Conclusions: Older, non-White (Black, Asian/Pacific Islander, American Indian/Alaska Native), Hispanic/Latino, unmarried men, residents of non-metropolitan areas, or those with lower annual household incomes have higher odds of late-stage diagnosis. Interventions targeting early detection in these at-risk groups may improve prostate cancer outcomes in the US. Variable Odds Ratio (95% CI) p-value Race (Black vs White) 1.24 (1.22-1.27) <0.01 Race (Asian or Pacific Islander vs White) 1.35 (1.31-1.40) <0.01 Race (American Indian/Alaska Native vs White) 1.70 (1.53-1.89) <0.01 Ethnicity (Hispanic/Latino vs non-/Hispanic/Latino) 1.45 (1.42-1.49) <0.01 Age (≥65 vs <65) 1.68 (1.65-1.71) <0.01 Marital status (unmarried vs married) 1.88 (1.85-1.91) <0.01 Residence status (non-metropolitan vs metropolitan) 1.10 (1.07-1.12) <0.01 Household income (<$70,000 vs ≥$70,000) 1.03 (1.01-1.05) <0.01

341 Background: Specialist clinicians (oncologists, urologists, radiation oncologists) read through free text reports of prostate MRI and biopsy reports to risk stratify patients during initial consultation of localized PCa. This laborious and time-consuming process is repeated for each day during the clinic. Herein, we propose a hybrid stratification framework utilizing a combination of large language model (LLM) and rule-based programming to automate this process. Methods: This retrospective study included patients with localized PCa (2004-2024) presenting to care at Mayo Clinic with at least one positive prostate biopsy and MRI report available. The state-of-the-art LLM – GPT4 – was utilized using a structured zeroshot prompt to extract relevant phenotypic variables (clinical T stage, Gleason patterns, grade group, extent of specimen/cores involved, prostate size/volume) from unstructured MRI and biopsy reports. Using a rule-based algorithm, the extracted variables and PSA at PCa diagnosis were used to categorize each patient into one of the six NCCN risk categories (very low [VL], low [L], favorable intermediate [FInt], unfavorable intermediate [UFInt], high risk (HR), very high risk [VHR]). Prompts were iteratively developed using ~5% of the total dataset and validated on ~10% of the dataset. Final performance was assessed against a held-out expert annotated test dataset using evaluation metrics (accuracy, F1-score, precision, recall). Additionally, manual annotation was conducted by two independent novice reviewers to compare machine annotated and novice-human annotated risk categorization. Results: A total of 397 patients were included in the evaluation. The median age at diagnosis was 64.8 (IQR: 59.7-68.5); majority of the men were White (n: 366; 92%) and non-Hispanic (n: 371; 94%). The most prevalent risk category was UFint (n: 151; 38%) followed by FInt (n: 80; 20.2%), HR (n: 76; 19.1%), VHR (n: 64; 16.1%), L (21: 5.3%), and VL (n: 5; 1.3%). Novice reviewers achieved an accuracy of 79%, precision of 80%, recall of 82% and a F1-score of 0.81. The hybrid stratification framework using GPT4 achieved an accuracy of 89%, precision of 89%, recall of 88% and a F1-score of 0.88. Among 43 errors by GPT4 risk stratification, majority of the errors were due to miscategorization of clinical T-stage (n: 21; 49%) followed by erroneous number of positive cores (n: 13; 30%). However, additional evaluation showed that GPT4 achieved numerically higher performance (accuracy: 90%; precision: 91%; recall: 90%; F1: 0.90) compared to novice reviewers (accuracy: 77%; precision: 89%; recall: 77%; F1: 0.82) for ascertaining clinical T-stage. Conclusions: Large language models exhibited superior performance than novice clinicians for risk stratification in patients with localized PCa. As the next step, we intend to validate our framework for automated risk classification in a prospective manner.

184 Background: We update evidence on mHSPC when new evidence becomes available (PMID: 36862387). ARANOTE trial prompted us to assess the comparative efficacy of darolutamide (DARO) with other ARPi agents in mHSPC using the living network meta-analysis (LNMA). Methods: This LNMA is conducted using the living interactive evidence (LIvE) synthesis framework. Phase III trials assessing systemic treatments in mHSPC are included. Mixed treatment comparisons were made using a frequentist network meta-analysis. P-scores were computed to assess relative treatment rankings with higher values indicating potentially better efficacy. Results: This report of LNMA includes a total 11 trials (12628 patients; 12 unique treatment options) as of October 1 st , 2024. Overall results were consistent with prior updates. Analysis limited to the ARPi trials showed that inhigh volume, abiraterone acetate (AAP)+androgen deprivation therapy (ADT) (HR: 0.46; 95% CI: 0.40-0.53; rank 1), apalutamide (APA)+ADT (0.53; 0.41-0.68; rank 3), DARO+ADT (0.60; 0.44-0.81; rank 4) and enzalutamide (E)+ADT (0.48; 0.41-0.56; rank 2) significantly improved radiographic progression-free survival (rPFS) compared to ADT alone. No statistically significant differences were observed with DARO+ADT compared to AAP+ADT (1.30; 0.94-1.82), APA+ADT (1.14; 0.77-1.67) and E+ADT (1.24; 0.89-1.79). Inlow volume, AAP+ADT (0.48; 0.37-0.63; rank 4), APA+ADT (0.36; 0.22-0.58; rank 3), DARO+ADT (0.30; 0.15-0.60; rank 1) and E+ADT (0.32; 0.25-0.41; rank 2) significantly improved rPFS compared to ADT. E+ADT significantly improved rPFS compared to AAP+ADT (0.67; 0.47-0.96) in low volume. No statistically significant differences were observed with DARO+ADT compared to AAP+ADT (0.62; 0.30-1.30), APA+ADT (0.83; 0.36-1.92) and E+ADT (0.93; 0.45-1.94). In synchronous disease, AAP+ADT (0.58; 0.51-0.67; rank 4), APA+ADT (0.49; 0.39-0.62; rank 2), DARO+ADT (0.59; 0.44-0.79; rank 3) and E+ADT (0.42; 0.36-0.50; rank 1) significantly improved rPFS compared to ADT. E+ADT significantly improved rPFS compared to AAP+ADT (0.73; 0.59-0.89) in synchronous disease. In metachronous disease, DARO+ADT (0.34; 0.17-0.67; rank 1), APA+ADT (0.41; 0.22-0.77; rank 2) and E+ADT (0.44; 0.35-0.57; rank 3) significantly improved rPFS compared to ADT. No statistically significant differences were observed with DARO+ADT compared to APA+ADT (0.83; 0.33-2.08) and E+ADT (0.76; 0.37-1.57) in metachronous disease. The results were consistent for overall survival. Conclusions: Current evidence suggests no significant differences for darolutamide as compared to other ARPi agents. Enzalutamide may be preferred over AAP in low-volume or synchronous disease. Given similar efficacy, choice of ARPi requires careful consideration of patient-specific factors including cost, accessibility and toxicity.

866 Background: Emerging data suggests that Language Models (LLMs) provide an unprecedented opportunity to extract critical data from the unstructured pathology reports. However, the performance of state of art (SOTA) models on clinically meaningful tasks such as assessment of pathologic responses in patients receiving neoadjuvant chemotherapy (NAC) in muscle invasive bladder cancer(MIBC) is not well known. Methods: This retrospective cohort included patients with pathologically confirmed Muscle-Invasive Bladder Cancer from the year 2018 to 2024. Selected patients included those who underwent Neoadjuvant chemotherapy followed by definitive cystectomy at Mayo Clinic. Gold standard labels were manually curated by trained clinicians. The state of the art (SOTA) LLM – GPT4 – was utilized using a structured zero-shot prompt to extract relevant phenotypic variables such as tumor site, tumor histology and additional histological variants if any, presence of in-situ / invasive carcinoma, presence of lamina propria / muscularis propria / lympho-vascular / perineural invasion, and T-stage from unstructured pathology reports. Structured prompts were iteratively developed for each data variable and validated using ~5% of the total dataset. The extracted variables were manually compared and labelled as True positive / True negative / False positive / False negative. Final performance was assessed against a held-out expert annotated test dataset using evaluation metric (accuracy). Results: A total of 200 reports from 99 patients were extracted by the LLM. Of the 200 reports, 41 duplicate / other pathology reports (e.g. – autopsy, cholecystectomy, etc.) were removed during final assessment. Significant characteristics such as tumor site, histology, and additional histology variant had an accuracy of 89%, 87%, and 91% respectively. Notably, accuracy of detection of in-situ versus invasive carcinoma was 70% versus 91%. Specific variables such as lamina propria / muscularis propria invasion / T-stage / lympho-vascular invasion / perineural invasion were answered with the accuracy of 75%, 75%, 67%, 90%, and 94% respectively. Conclusions: LLMs have the potential to extract critical data from unstructured pathology reports and automate assessment of pathologic responses in MIBC patients receiving NAC. Iterative improvements in performance can be achieved by improving prompt design using expert clinical input. This process can be potentially scaled to automate cancer registries.

Abstract Purpose: Data on clear-cell renal cell carcinoma (ccRCC) xenografts defined the seleno-L-methionine (SLM) dose and the plasma selenium concentrations associated with the enhancement of hypoxia-inducible factor-1α/2α degradation, stabilization of tumor vasculature, enhanced drug delivery, and efficacy of axitinib. The data provided the rationale for the development of this phase I clinical trial of SLM and axitinib in advanced or metastatic relapsed ccRCC. Patients and Methods: Patients were ≥18 years with histologically and radiologically confirmed advanced or metastatic ccRCC who had received at least one prior systemic therapy, which could include axitinib (last dose ≥6 months prior to enrollment). Escalating dose levels of SLM (2,500, 3,000, and 4,000 μg) were administered orally twice daily for 14 days and then once daily concurrently with axitinib 5 mg twice daily using a 3 + 3 design in phase I. Patients were treated at the 4,000 μg dose level in the expansion cohort to obtain preliminary estimates of efficacy. Results: No dose-limiting toxicities occurred at the 4,000 μg SLM dose level. Among the 27 patients treated with 4,000 μg of SLM, the overall response rate was 55.6%, median duration of response was 18.4 months, median progression-free survival was 14.8 months, and median overall survival was 19.6 months. Preliminary results have shown that plasma selenium concentrations, inhibition of TGF-β1, and stabilization of tumor vasculature by SLM are time dependent. Conclusions: SLM (4,000 μg) in sequential combination with axitinib is well tolerated with encouraging efficacy.

385 Background: Tyrosine kinase inhibitors (TKI) and immuno-oncology (IO) agents alone or in combination have revolutionized the landscape of management for untreated metastatic renal cell carcinoma (mRCC). Baseline quality of life (QOL) outside of interventional clinical trials and rationale for management decisions in mRCC are poorly understood. Methods: In this prospective observational study of 800 mRCC pts in the US, pts must: have a diagnosis of mRCC (any histology) with no prior systemic therapy (ST) for mRCC; be age ≥19 at informed consent; be able to comply with completion of PROs. Exclusion criteria include being treated for active malignancy other than mRCC or not intending to follow up at a study site within PCORnet. Pts undergo consent and baseline assessments by the study site team, with subsequent follow up by the central coordinating center. The primary objective is to determine distinct patterns of change in QOL and symptom burden of mRCC pts. A key secondary objective is to identify patterns of clinical management in the real world setting of mRCC across treatment regimens. Minimally important differences are defined as 3- and 7-points for FKSI-19 and FACT-G respectively. ClinicalTrials.gov ID: NCT04919122. Results: As of 9/1/2023, 173 pts have been enrolled (126 ST, 47 no ST [NST]): 75% male, 87% white, 73% clear cell, 52% intermediate risk and 47% poor risk. 47% had radical nephrectomy and 22% another solid tumor. ST pts as compared to NST pts were more likely to be poor risk (52% vs 16%); but differences in baseline pain were not significant. Of ST pts, 44 were treated with IO-IO (ipilimumab + nivolumab), 50 with IO-TKI (cabozantinib + nivolumab 38%, axitinib + pembrolizumab 38%, lenvatinib + pembrolizumab 24%) and 32 other (IO alone 34%, TKI alone 31%, investigational drug 25%). Reasons for NST included active surveillance (AS, 50%), other (21%), local therapy (13%). Mean differences between ST and NST for baseline FKSI-19 and FACT-G were 5.7 (95% CI: 1.2 – 10.2) and 3.3 (-2.6 – 9.3). There were no differences in pain or PROs between IO-IO and IO-TKI pts. Mean differences in FKSI-19 and FACT-G between AS vs IO-IO were 10.6 (4.8 – 16.4) and 8.6 (1.1 – 16.0); and vs IO-TKI were 11.0 (4.9 – 17.0) and 10.3 (3.5 – 17.1). On physician surveys, primary reasons for treatment selection were (in order of % selected): complete response (CR), prolonged survival (OS), prognostic factors (Px), treatment-free interval for IO-IO; versus delaying progression (PFS), OS, CR, Px for IO-TKI. Conclusions: Our data suggest differences in baseline QoL in pts treated with ST vs NST, as well as AS vs IO-IO or IO-TKI, in the real world. Reasons for selection of ST regimen differed among combination regimen classes. Completion of ODYSSEY will clarify baseline and longitudinal differences in pain and PROs among management strategies and give novel insights into the rationale for real world management of untreated pts with mRCC. Clinical trial information: NCT04919122 .

ImportanceSystematic data on the association between anticancer therapies and thromboembolic events (TEEs) in patients with COVID-19 are lacking.ObjectiveTo assess the association between anticancer therapy exposure within 3 months prior to COVID-19 and TEEs following COVID-19 diagnosis in patients with cancer.Design, Setting, and ParticipantsThis registry-based retrospective cohort study included patients who were hospitalized and had active cancer and laboratory-confirmed SARS-CoV-2 infection. Data were accrued from March 2020 to December 2021 and analyzed from December 2021 to October 2022.ExposureTreatments of interest (TOIs) (endocrine therapy, vascular endothelial growth factor inhibitors/tyrosine kinase inhibitors [VEGFis/TKIs], immunomodulators [IMiDs], immune checkpoint inhibitors [ICIs], chemotherapy) vs reference (no systemic therapy) in 3 months prior to COVID-19.Main Outcomes and MeasuresMain outcomes were (1) venous thromboembolism (VTE) and (2) arterial thromboembolism (ATE). Secondary outcome was severity of COVID-19 (rates of intensive care unit admission, mechanical ventilation, 30-day all-cause mortality following TEEs in TOI vs reference group) at 30-day follow-up.ResultsOf 4988 hospitalized patients with cancer (median [IQR] age, 69 [59-78] years; 2608 [52%] male), 1869 had received 1 or more TOIs. Incidence of VTE was higher in all TOI groups: endocrine therapy, 7%; VEGFis/TKIs, 10%; IMiDs, 8%; ICIs, 12%; and chemotherapy, 10%, compared with patients not receiving systemic therapies (6%). In multivariable log-binomial regression analyses, relative risk of VTE (adjusted risk ratio [aRR], 1.33; 95% CI, 1.04-1.69) but not ATE (aRR, 0.81; 95% CI, 0.56-1.16) was significantly higher in those exposed to all TOIs pooled together vs those with no exposure. Among individual drugs, ICIs were significantly associated with VTE (aRR, 1.45; 95% CI, 1.01-2.07). Also noted were significant associations between VTE and active and progressing cancer (aRR, 1.43; 95% CI, 1.01-2.03), history of VTE (aRR, 3.10; 95% CI, 2.38-4.04), and high-risk site of cancer (aRR, 1.42; 95% CI, 1.14-1.75). Black patients had a higher risk of TEEs (aRR, 1.24; 95% CI, 1.03-1.50) than White patients. Patients with TEEs had high intensive care unit admission (46%) and mechanical ventilation (31%) rates. Relative risk of death in patients with TEEs was higher in those exposed to TOIs vs not (aRR, 1.12; 95% CI, 0.91-1.38) and was significantly associated with poor performance status (aRR, 1.77; 95% CI, 1.30-2.40) and active/progressing cancer (aRR, 1.55; 95% CI, 1.13-2.13).Conclusions and RelevanceIn this cohort study, relative risk of developing VTE was high among patients receiving TOIs and varied by the type of therapy, underlying risk factors, and demographics, such as race and ethnicity. These findings highlight the need for close monitoring and perhaps personalized thromboprophylaxis to prevent morbidity and mortality associated with COVID-19–related thromboembolism in patients with cancer.

Abstract Purpose: Mucosal melanoma is a rare, aggressive form of melanoma with extremely high recurrence rates despite definitive surgical resection with curative intent. Currently there is no consensus on adjuvant therapy. Data on checkpoint inhibitors for adjuvant therapy are lacking. Patients and Methods: We performed a single-arm, multicenter clinical trial using “flip dose” ipilimumab (1 mg/kg q3w × 4 cycles), and nivolumab (3 mg/kg q3w × 4 cycles), then nivolumab 480 mg q4w × 11 cycles to complete a year of adjuvant therapy. Participants must have had R0/R1 resection ≤90 days before registration, no prior systemic therapy (adjuvant radiotherapy allowed), ECOG 0/1, and no uncontrolled autoimmune disease or other invasive cancer. Patients were recruited through the Midwest Melanoma Partnership/Hoosier Oncology Network. Results: From September 2017 to August 2021, 35 patients were enrolled. Of these, 29 (83%) had R0 resections, and 7 (20%) received adjuvant radiotherapy. Median age was 67 years, 21 (60.0%) female. Recurrence-free survival (RFS) rates at 1 and 2 years were 50% [95% confidence interval (CI), 31%–66%] and 37% (95% CI, 19%–55%), respectively. Overall survival rates at 1 and 2 years were 87% (95% CI, 68%–95%) and 68% (95% CI, 46%–83%), respectively. Median RFS was 10.3 months (95% CI, 5.7–25.8). Most common grade 3 toxicities were diarrhea (14%), hypertension (14%), and hyponatremia (11%), with no grade 4/5 toxicities. Conclusions: Flip-dose ipilimumab and nivolumab after resection of mucosal melanoma is associated with outcomes improved over that of surgical resection alone. Long-term follow-up, subgroup analyses and correlative studies are ongoing.