Citations

Background Peripheral T-cell lymphomas (PTCLs) comprise a rare, heterogeneous group of non-Hodgkin lymphomas of mature T-cell origin. Most systemic (s) PTCL subtypes typically have poor outcomes with conventional anthracycline-based chemotherapy. Biomarker-driven and subtype-specific treatments are promising but unmet needs. Few benchmark studies exist to characterize the current real-world landscape of clinical and pathologic practice. We report the patterns of care and updated outcome data for PTCL patients enrolled in the LEO-MER multi-center prospective cohort study (ClinicalTrials.gov NCT02736357). Methods Patients aged 18 years or older with newly diagnosed PTCL were prospectively enrolled in the University of Iowa/Mayo Clinic MER cohort (2002-2015) or the expanded LEO cohort (2015-2020). Clinical, pathology, treatment, and outcome data were abstracted from medical records using a standard protocol in both cohorts. Pathology underwent expert re-review based on WHO criteria. Overall survival (OS) was calculated from date of diagnosis to date of death or last follow-up. Event-free survival (EFS) was calculated from date of diagnosis to disease progression, initiation of 2 nd line therapy, or death from any cause. EFS and OS were evaluated using Cox model and Kaplan-Meier estimator. Log-rank test was used to test the significance of difference between groups. Results LEO-MER enrolled 1132 PTCL patients during the study period (462 MER and 670 LEO). Of these, 722 were sPTCLs (281 MER and 441 LEO), including PTCL-NOS (N=253, 35%), anaplastic large cell lymphoma (ALCL; N=180, 24.9% [74 ALK+, 106 ALK-]), angioimmunoblastic T-cell lymphoma (AITL, N=164, 22.7%), extranodal NK-TCL (N=50, 6.9%), adult T-cell leukemia/lymphoma (N=33, 4.6%), enteropathy-associated TCL (N=19, 2.6%), and hepatosplenic TCL (N=13, 1.8%) ( Table 1). Median age at diagnosis was 60 years. M:F ratio was 1.5. Comparing LEO to MER, African American population was 15.4% vs 0.7%, and Hispanic was 10% vs 2.2%. Clinical characteristics were comparable between MER and LEO overall: PS≥2, 20.9%; elevated LDH, 50.7%; stage III-IV, 67.6%; and IPI 2-5, 65.7%. The most common 1 st line chemotherapy regimens overall were CHOP-based (N=506, 70%), including 60% receiving CHOP-like chemotherapy in both MER and LEO cohorts (CHOP [N=268, 37%], CHOEP [N=91, 13%] or EPOCH [N=75, 10%]), and 16% with CHOP-like in combination with novel agents in LEO cohort (BV+ [N=41, 9.3%], azacitidine+ [N=11, 2.5%], pralatrexate+ [N=10, 2.3%], lenalidomide+ [N=8, 1.9%]) ( Table 1). More patients received etoposide in LEO (40.3%) than in MER (20.3%). 74 patients (10.3%) underwent consolidative SCT following induction chemotherapy (71 auto and 3 allo). 65 patients (11.2%) received frontline therapy on clinical trials. At a median follow-up of 9.4 years in MER and 4 years in LEO, 177 (MER) and 269 (LEO) events, and 150 (MER) and 206 (LEO) deaths were observed, respectively. The 3-year EFS and 4-year OS estimates for sPTCL were 46.3% and 59.1% in MER, and 40.4% and 51.3% in LEO. EFS and OS correlated with IPI scores and differed by sPTCL subtypes ( Figure 1). ALCL had superior survival, with 3-year EFS and 4-year OS of 78.2% and 90.5% for ALK+, and 65.9% and 76.9% for ALK-. Non-ALCL subtypes including PTCL-NOS and AITL had inferior outcomes, with 3-year EFS and 4-year OS of 31.4% and 42.8% for PTCL-NOS, and 33.4% and 50.7% for AITL. In subset analysis, the addition of etoposide to CHOP chemotherapy as either CHOEP or EPOCH was associated with improved OS for ALK- ALCL (p=0.038, log-rank), but not for ALK+ ALCL (p=0.75) or non-ALCL subtypes of PTCL-NOS and AITL (p=0.88). Conclusion The LEO-MER study is the largest prospective cohort study of PTCL to date. Patterns of care in the LEO cohort begin to incorporate novel agents in frontline therapy. Outcomes continue to mature with longitudinal follow-up and ongoing accrual, which poise to shape benchmarks in the contemporary era. The lack of benefit of etoposide adding to CHOP induction and poor overall survival of non-ALCL subtypes underscores the unmet need of therapeutic breakthrough for non-ALCL frontline treatment, particularly through clinical trials with biomarker-guided approaches incorporating novel agents.

INTRODUCTION: Inadequate bowel preparation can result in decreased diagnostic accuracy and therapeutic safety of colonoscopy for colon cancer screening. The Boston Bowel Preparation Scale (BBPS) has been used to assess the quality of bowel preparation. Glucagon-like peptide-1 receptor agonists (GLP-1RA) are commonly used medications for diabetes mellitus and obesity that are known to delay gastrointestinal motility. We hypothesized that the use of GLP-1RA would be associated with decreased quality of bowel preparation. METHODS: We performed a retrospective cohort study of patients who underwent screening or surveillance colonoscopy at a large academic medical center between December 2021 and December 2022. We included patients taking any GLP-1RA for diabetes or obesity during colonoscopy defined as our cases, and patients who were prescribed GLP-1RA at one point but not within 3 months of colonoscopy defined as our controls. We excluded patients on any promotility or antimotility agents and those without BBPS recorded on their procedure report. Independent t test assessed statistical differences in the case and control groups to compare the quality of bowel preparation for continuous variables, and the χ 2 test was used for categorical variables. Multivariate linear regression including diabetes as a covariate was also performed for continuous variables, and multivariate logistic regression was performed for categorical variables. RESULTS: A total of 446 patients were included in the study, comprising 265 (59%) cases and 181 controls (41%). There were no statistically significant differences between groups at baseline except for the diagnosis of diabetes ( P = 0.001) with a higher proportion of patients with diabetes in the cases. The mean BBPS was significantly higher in controls (7.0 ± 1.9 vs 7.5 ± 2.4, P = 0.046) when controlling for diabetes. The percentage of patients with a total BBPS score of <5 was significantly higher in cases (15.5% vs 6.6%, P = 0.01). The proportion of patients who required a repeat colonoscopy due to poor bowel preparation was also significantly higher in cases (18.9% vs 11.1%, P = 0.041). DISCUSSION: The use of GLP-1RA was associated with a statistically significantly lower quality of bowel preparation, with additional clinical significance given a notable difference in the need for a repeat colonoscopy. It will be essential to understand the cumulative effect of medications that may delay gastric emptying on the quality of bowel preparation to better understand the appropriate measures and counseling that need to be taken before undergoing outpatient colonoscopies.