Machine Learning-Based Survival Prediction Models for Progression-Free and Overall Survival in Advanced-Stage Hodgkin Lymphoma.

PURPOSE: Patients diagnosed with advanced-stage Hodgkin lymphoma (aHL) have historically been risk-stratified using the International Prognostic Score (IPS). This study investigated if a machine learning (ML) approach could outperform existing models when it comes to predicting overall survival (OS) and progression-free survival (PFS). PATIENTS AND METHODS: This study used patient data from the Danish National Lymphoma Register for model development (development cohort). The ML model was developed using stacking, which combines several predictive survival models (Cox proportional hazard, flexible parametric model, IPS, principal component, penalized regression) into a single model, and was compared with two versions of IPS (IPS-3 and IPS-7) and the newly developed aHL international prognostic index (A-HIPI). Internal model validation was performed using nested cross-validation, and external validation was performed using patient data from the Swedish Lymphoma Register and Cancer Registry of Norway (validation cohort). RESULTS: In total, 707 and 760 patients with aHL were included in the development and validation cohorts, respectively. Examining model performance for OS in the development cohort, the concordance index (C-index) for the ML model, IPS-7, IPS-3, and A-HIPI was found to be 0.789, 0.608, 0.650, and 0.768, respectively. The corresponding estimates in the validation cohort were 0.749, 0.700, 0.663, and 0.741. For PFS, the ML model achieved the highest C-index in both cohorts (0.665 in the development cohort and 0.691 in the validation cohort). The time-varying AUCs for both the ML model and the A-HIPI were consistently higher in both cohorts compared with the IPS models within the first 5 years after diagnosis. CONCLUSION: The new prognostic model for aHL on the basis of ML techniques demonstrated a substantial improvement compared with the IPS models, but yielded a limited improvement in predictive performance compared with the A-HIPI.

Use of statins and risk of myeloproliferative neoplasms: a Danish nationwide case-control study.

Previous studies have indicated a possible cancer-protective effect of statins in solid cancers; however, this has never been investigated in myeloproliferative neoplasms (MPNs). We aimed to investigate the association between statin use and the risk of MPNs in a nested nationwide case-control study, using Danish national population registries. Information on statin use was obtained from the Danish National Prescription Registry, and patients diagnosed with MPNs between 2010 and 2018 were identified from the Danish National Chronic Myeloid Neoplasia Registry. The association between statin use and MPNs was estimated using age- and sex-adjusted odds ratios (ORs) and fully adjusted ORs (aORs), adjusting for prespecified confounders. The study population included 3816 cases with MPNs and 19 080 population controls (5:1) matched for age and sex using incidence density sampling. Overall, 34.9% of the cases and 33.5% of the controls ever used statins, resulting in an OR for MPN of 1.07 (95% confidence interval [CI], 0.99-1.16) and an aOR of 0.87 (95% CI, 0.80-0.96), respectively. 17.2% were categorized as long-term users (≥5 years) among the cases compared with 19.0% among controls, yielding an OR for MPN of 0.90 (95% CI, 0.81-1.00) and an aOR of 0.72 (95% CI, 0.64-0.81). Analysis of the effect of the cumulative duration of statin use revealed a dose-dependent response, and the association was consistent for sex, age, and MPN subgroups and across different statin types. Statin users were associated with significantly lower odds of being diagnosed with an MPN, indicating a possible cancer-preventive effect of statins. The retrospective of this study precludes causal inferences.

Outcome of limited-stage peripheral T-Cell lymphoma after CHOP(-like) therapy: A population based study of 239 patients from the Nordic lymphoma epidemiology group.

Peripheral T-Cell Lymphomas (PTCLs) are rare, aggressive lymphomas with poor outcomes, but limited-stage disease is infrequent and not well-described. This study reports outcomes and prognostic factors in limited-stage nodal PTCLs in a binational population-based setting. Patients were identified from the Danish and Swedish lymphoma registries. Adults diagnosed with limited-stage nodal PTCL (stage I-II) and treated with CHOP(-like) therapy ±radiotherapy between 2000 and 2014 were included. Medical records were reviewed by local investigators. A total of 239 patients with a median age of 62 years were included; 67% received 6-8 cycles of CHOP(-like) therapy and 22% received 3-4 cycles, of which 59% also received radiotherapy. Autologous stem cell transplant consolidation was administered to 16% of all patients. Median follow-up was 127 months with 5-years overall survival (OS) of 58% (95% CI: 53-65) and progression-free survival (PFS) of 53% (95% CI: 47-59). In multivariable analysis, age ≥ 60 years and B-symptoms were unfavorable and ALK+ anaplastic large cell T-Cell lymphoma was favorable for survival outcomes. There was no difference in treatment-specific outcome (3-4 cycles vs. 6-8 cycles of CHOP(-like) ± radiotherapy). Low-risk patients (age < 60 without B-symptoms) had a 5-year OS of 77% (95% CI 67-89%). In the present study of limited-stage nodal PTCL, survival after curative intent chemotherapy +/- radiotherapy was inferior to that of limited-stage diffuse large B-cell lymphoma, but a subgroup of young patients without B-symptoms had very good outcomes. Treatment outcomes after 3-4 cycles versus 6-8 cycles of CHOP(-like) therapy were comparable.

Temporal changes in survival among adult patients with acute myeloid leukaemia in the period 2000-2016: a Danish population-based study.

In the present study, we quantify the progress in overall survival (OS) during the period 2000-2016 among Danish patients with acute myeloid leukaemia (AML). This population-based study, including 3820 adult patients with AML, demonstrates a significantly improved OS over time with the 2-year age-standardised OS increasing from 22% in 2002 to 31% in 2016. The improvement in OS was exclusively seen in patients with AML aged ≥50 years, with absolute improvements in 2-year OS from 2002 to 2016 of ≥10% among patients aged 50-75 years and a small absolute increase in those aged >75 years.