Citations

PURPOSE Cholangiocarcinoma is a rare cancer associated with poor prognosis. Pemigatinib was the first FGFR1-3 inhibitor approved for second-line therapy and beyond, on the basis of the phase II FIGHT-202 trial. Here, we assess efficacy and safety of first-line pemigatinib. METHODS FIGHT-302 is a phase III, randomized, global trial evaluating pemigatinib as first-line therapy (ClinicalTrials.gov identifier: NCT03656536 ). Adults with advanced cholangiocarcinoma with FGFR2 rearrangement were randomized 1:1 to receive pemigatinib (13.5 mg once daily) or chemotherapy (1,000 mg/m 2 gemcitabine plus 25 mg/m 2 cisplatin once per day on days 1 and 8 of every 3-week cycle for ≤8 cycles) and were stratified by prior receipt of chemotherapy, geographic region, and tumor burden. Pemigatinib crossover was allowed for patients progressing on chemotherapy. The primary end point was progression-free survival (PFS). Secondary efficacy, safety, and exploratory end points were also analyzed. RESULTS Overall, 4,563 patients were prescreened, 196 were screened, and 167 randomly assigned to receive pemigatinib (n = 83) or chemotherapy (n = 84) before early closure of the study because of a change in standard of care. The median PFS was 8.3 months in the pemigatinib group versus 6.8 months in the chemotherapy group (hazard ratio, 0.58 [95% CI, 0.39 to 0.87]; nominal P = .0078); objective response rate was 47% versus 15%, and the median duration of response was 14.2 versus 6.3 months. Median overall survival was similar (24.4 v 25.0 months, respectively). In the crossover group (n = 42, second-line pemigatinib), the median PFS was 8.1 months. Safety was consistent with the known profile of pemigatinib. CONCLUSION To our knowledge, this was the largest, first-line, randomized, phase III trial of a targeted therapy for advanced FGFR2- rearranged cholangiocarcinoma. Pemigatinib demonstrated prolonged median PFS compared with chemotherapy, with no new safety findings.

4017 Background: First-line mCCA treatment is chemotherapy and immunotherapy. Based on the phase 2 FIGHT-202 trial, pemigatinib was the first FGFR1-3 inhibitor approved in second line and beyond for mCCA with FGFR2 rearrangement. Results from the phase 3, randomized, global FIGHT-302 trial evaluating pemigatinib as first-line therapy are reported here (NCT03656536). Methods: Patients (pts) aged ≥18 y with previously untreated mCCA with FGFR2 rearrangement were randomized 1:1 to receive oral pemigatinib (13.5 mg once daily q21) or intravenous chemotherapy (1000 mg/m 2 gemcitabine + 25 mg/m 2 cisplatin d1,8 q21 ≤8 cycles) stratified by previous receipt of 1 chemotherapy cycle, geographic region, and tumor burden. Pemigatinib crossover was allowed for control arm pts progressing on chemotherapy. Primary end point was progression-free survival (PFS). Secondary efficacy, safety, and exploratory end points were also analyzed. Results: After prescreening >4000 pts, only 196 had FGFR2 rearrangement and were screened; 167 were randomized to receive pemigatinib (n=83) or chemotherapy (n=84) before closing the study early due to slow accrual. Baseline characteristics and demographics were similar between arms. Median PFS (mPFS) with pemigatinib versus chemotherapy was 8.3 versus 6.8 mo, respectively (hazard ratio [95% CI], 0.58 [0.39-0.87]; nominal P =.0078); secondary efficacy endpoints generally supported primary results (Table). Median OS was similar, although these results are confounded by FGFR inhibitor administration as second-line treatment. In the crossover group (n=42, second-line pemigatinib), mPFS was 8.1 mo. Safety was consistent with previous results (Table). Exploratory analyses of genetic co-alterations associated with treatment response will be presented. Conclusions: FIGHT-302 was the first phase 3 trial of a targeted therapy for untreated mCCA. Pemigatinb demonstrated superior activity and prolonged mPFS compared with first-line chemotherapy. The results confirm the current management paradigm of pemigatinib administration after disease progression on chemotherapy. Clinical trial information: NCT03656536 . Pemigatinib Gemcitabine + cisplatin mPFS (95% CI), mo a 8.3 (6.5-12.2) 6.8 (6.1-8.3) Objective response rate, n/N (%) a 39/83 (47) 13/84 (15.5) Median duration of response (95% CI), mo a 14.2 (8.7-24.7) 6.3 (4.3-not estimable) Median overall survival (95% CI) a,b , mo 24.4 (18.6-35.9) 25.0 (18.7-34.2) Treatment related TEAEs, n/N c (%) 80/83 (96) 70/73 (96) Serious TEAEs, n/N (%) c,d 23/83 (28) 17/73 (23) Grade ≥3 TEAEs, n/N (%) c 65/83 (78) 49/73 (67) TEAEs leading to treatment discontinuation, n/N (%) c 5/83 (6) 8/73 (11) Fatal TEAE, n/N (%) c,d 3/83 (4) 0 a All randomized pts. b Interpretation limited by second-line treatment with FGFR inhibitor in gemcitabine + cisplatin arm. c Pts who received ≥1 study dose. d None deemed treatment related.

ABSTRACT Background Systemic infections are a leading cause of hospitalization and death among patients with cirrhosis. Timely and accurate infection identification is essential for both clinical care and the development of predictive models. However, existing methods such as ICD-10 coding are unreliable, and manual chart review is resource-intensive and difficult to scale. This study aimed to develop and validate an automated large language model (LLM)-based approach for infection classification and subtyping in patients with cirrhosis presenting to the emergency department (ED). Method We developed INFEHR (INfection identification and subtyping using Free-text EHR analysis), an LLM-powered pipeline utilizing Claude 3.5 Sonnet to analyze clinical notes from the first 72 hours of admission. Model outputs were compared against a physician-adjudicated gold standard in a cohort of 1,000 encounters from patients with cirrhosis who presented to the ED. Performance was benchmarked against ICD-10 code–based labeling and CDC Adult Sepsis Event criteria. Results INFEHR achieved 94.7% overall accuracy, with 99.5% sensitivity and 92.8% positive predictive value for identifying infection presence, outperforming ICD-10–based classification across all metrics ( p < 0.0001). The model also demonstrated strong performance in classifying pathogen type and infection site. This pipeline processed notes within seconds, offering improvements in efficiency and scalability over manual review. Conclusion INFEHR offers a scalable, reproducible, and accurate method for infection phenotyping in cirrhosis. By overcoming limitations of traditional coding and manual review, it supports high-throughput infection surveillance, improves cohort construction for clinical research, and enables future integration into real-time decision-support tools in hepatology.

PURPOSE Manual extraction of treatment outcomes from unstructured oncology clinical notes is a significant challenge for real-world evidence (RWE) generation. This study aimed to develop and evaluate a robust natural language processing (NLP) system to automatically extract cancer treatments and their associated RECIST-based response categories (complete response, partial response, stable disease, and progressive disease) from non–small cell lung cancer (NSCLC) clinical notes. METHODS This retrospective NLP development and validation study used a corpus of 250 NSCLC oncology notes from University of Pittsburgh Medical Center (UPMC) Hillman Cancer Center, annotated by physician experts. An end-to-end NLP pipeline was designed, integrating a rule-based module for entity extraction (treatments and responses) and a machine learning module using biomedical clinical bidirectional encoder representations from transformers for relation classification. The system's performance was evaluated on a held-out test set, with partial external validation for relation extraction on a Mayo Clinic data set. RESULTS The NLP system achieved high overall accuracy. On the UPMC test set (64 notes), the relation classification model attained an area under the receiver operating characteristic curve of 0.94 and an F1 score of 0.92 for linking treatments with documented responses. The rule-based entity extraction demonstrated a macro-averaged F1 score of 0.87 (precision 0.98, recall 0.81). Although precision was high for chemotherapy and most response types (1.00), recall for cancer surgery was 0.45. External validation at Mayo Clinic showed moderate relation extraction F1 scores (range: 0.51-0.64). CONCLUSION The proposed NLP system can reliably extract structured treatment and response information from unstructured NSCLC oncology notes with high accuracy. This automated approach can assist in abstracting critical cancer treatment outcomes from clinical narrative text, thereby streamlining real-world data analysis and supporting the generation of RWE in oncology.

Abstract The complementarity and clinical utility of combining liquid biopsies and radiomic image analysis has not been demonstrated. ctDNA minimal residual disease after chemoradiotherapy (CRT) for non–small cell lung cancer (NSCLC) is highly prognostic, but on-treatment biomarkers are needed to enable response-adapted therapies. In this study, we analyzed 418 patients with NSCLC undergoing CRT to develop and validate a novel dynamic risk model that accurately predicts ultimate progression-free survival during treatment. We optimize tissue-free variant calling from plasma samples to facilitate ctDNA monitoring and demonstrate the importance of accounting for persistent clonal hematopoiesis variants. We show that mid-CRT ctDNA concentration is prognostic for disease progression and integrate additional pre-CRT risk factors, including radiomics, into a combined model that improves outcome prediction. Our results suggest that tumor features, radiomics, and mid-CRT ctDNA analysis are complementary and can identify patients at high and low risk of progression to potentially enable response-adapted therapies. Significance: This study demonstrates that combining tumor features, radiomics, and ctDNA analysis improves outcome prediction in NSCLC treated with CRT therapy. Our integrated model could enable personalized and response-adapted therapies to reduce toxicity and improve outcomes in patients. See related commentary by Anagnostou and Aggarwal, p. 1534

AbstractProper cell‐cycle progression is essential for the self‐renewal and differentiation of human pluripotent stem cells (hPSCs). The fluorescent ubiquitination‐based cell‐cycle indicator (FUCCI) has allowed the dual‐color visualization of the G1 and S/G2/M phases in various dynamic models, but its application in hPSCs is not widely reported. In addition, lineage‐specific FUCCI reporters have not yet been developed to analyze complex tissue‐specific cell‐cycle progression during hPSC differentiation. Desiring a robust tool for spatiotemporal reporting of cell‐cycle events in hPSCs, we employed the CRISPR/Cas9 genome editing tool and successfully knocked the FUCCI reporter into the AAVS1 safe harbor locus of hPSCs for stable and constitutive FUCCI expression, exhibiting reliable cell‐cycle‐dependent fluorescence in both hPSCs and their differentiated progeny. We also established a cardiac‐specific TNNT2‐FUCCI reporter for lineage‐specific cell‐cycle monitoring of cardiomyocyte differentiation from hPSCs. This powerful and modular FUCCI system should provide numerous opportunities for studying human cell‐cycle activity, and enable the identification and investigation of novel regulators for adult tissue regeneration.

LncRNAs play an indispensable role in the process of M1 macrophage via regulating the development of macrophages and their responses to bacterial pathogens and viral infections. However, there are few studies on the lncRNA-mediated functions and regulatory mechanisms of M2 macrophage polarization. In this study, we found a number of differentially expressed lncRNAs between human monocyte derived M0 and M2 macrophages according to array analysis and qRT-PCR validation. The lncRNA RP11-389C8.2 (we named lnc-M2 in this study) was observed to be highly expressed in M2 macrophages. In Situ Localization and Quantification Analysis showed that lnc-M2 was expressed in the nucleus and cytosolic compartments of M2 macrophages. Notably, lnc-M2 knockdown enhanced the phagocytic ability of M2 macrophages. Ulteriorly, the results of RNA-Protein interaction experiments indicated that protein kinase A (PKA) was a lnc-M2 associated RNA-binding protein (RBP). Western blot showed that p-CREB, a well-known key downstream transcription factor of PKA, was lowly phosphorylated in lnc-M2-silencing M2 macrophages. Furthermore, we found that transcriptional factor STAT3 promoted lnc-M2 transcription along with the up-regulation of epigenetic histone modification markers at the lnc-M2 promoter locus, indicating that STAT3 activated lnc-M2 and eventually facilitated the process of M2 macrophage differentiation via the PKA/CREB pathway. Collectively, our date provide evidence that the transcription factor STAT3 can promote the transcription of lnc-M2 and facilitated the process of M2 macrophage differentiation via the PKA/CREB pathway. This study highlights a novel mechanism underlying the M2 macrophage differentiation.