RESUMO
MOTIVATION: A gradient boosting decision tree (GBDT) is a powerful ensemble machine-learning method that has the potential to accelerate biomarker discovery from high-dimensional molecular data. Recent algorithmic advances, such as extreme gradient boosting (XGB) and light gradient boosting (LGB), have rendered the GBDT training more efficient, scalable and accurate. However, these modern techniques have not yet been widely adopted in discovering biomarkers for censored survival outcomes, which are key clinical outcomes or endpoints in cancer studies. RESULTS: In this paper, we present a new R package 'Xsurv' as an integrated solution that applies two modern GBDT training frameworks namely, XGB and LGB, for the modeling of right-censored survival outcomes. Based on our simulations, we benchmark the new approaches against traditional methods including the stepwise Cox regression model and the original gradient boosting function implemented in the package 'gbm'. We also demonstrate the application of Xsurv in analyzing a melanoma methylation dataset. Together, these results suggest that Xsurv is a useful and computationally viable tool for screening a large number of prognostic candidate biomarkers, which may facilitate future translational and clinical research. AVAILABILITY AND IMPLEMENTATION: 'Xsurv' is freely available as an R package at: https://github.com/topycyao/Xsurv. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
Assuntos
Melanoma , Humanos , Prognóstico , Modelos de Riscos Proporcionais , BiomarcadoresRESUMO
Background: Standardized, high-quality PRO data reporting is crucial for patient centered care in the field of oncology, especially in clinical trials that establish standard of care. This study evaluated PRO endpoint design, conduct and reporting methods in FDA approved drugs for GU malignancies. Methods: A systematic review of the FDA archives identified GU cancer drug approvals from Feb 2007 to July 2022. ClinicalTrials.gov and PubMed were used to retrieve relevant data. PRO data was screened, and analytic tools, interpretation methods in the published papers and study protocols were reviewed. Compliance with PRO reporting standards were assessed using PRO Endpoint Analysis Score (PROEAS), a 24-point scoring scale from Setting International Standards in Analyzing Patient-Reported Outcomes and Quality of Life Endpoints Data Consortium (SISAQOL). Findings: We assessed 40 trial protocols with 27,011 participants, resulting in 14 renal cell cancer (RCC), 16 prostate cancer (PC), and 10 urothelial cancer (UC) approvals. PRO data was published for 27 trials, with 23 PRO publications (85%) focusing solely on PRO data, while 4 (15%) included PRO data in the original paper. Median time between primary clinical and secondary paper with PRO data was 10.5 months (range: 9-25 months). PROs were not planned as primary endpoints for any study but 14 (52%) reported them as secondary, 10 (37%) as exploratory outcomes, and 3 (11%) lacked any clarity on PRO data as endpoint. Mean PROEAS score of all GU cancers was 11.10 (range: 6-15), RCC (11.86, range: 6-15), UC (11.50, range: 9-14), and PC (10.56, range: 6-15). None met all the SISAQOL recommendations. Interpretation: Low overall PROEAS score and delays in PRO data publication in GU cancer drug trials conducted in the past decade emphasize the need for improvement in quality of design and conduct of PRO endpoint in future trials and accelerated publication of PRO endpoints, using standardized analysis, and prespecified hypothesis driven endpoint. These improvements are essential for facilitating interpretation and application of PRO study findings to define patient care. Funding: None.
RESUMO
Fludarabine (Flu) and melphalan (Mel) reduced-intensity conditioning is frequently used for allogenic hematopoietic cell transplant (allo-HCT) in patients with acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS). However, there is limited evidence on the impact of Mel dosing on toxicities and clinical outcomes of allo-HCT. We retrospectively compared 8/8 HLA-matched donor allo-HCT outcomes of 345 patients with AML or MDS receiving total Mel dose of 100 mg/m2 (Mel-100, nâ¯=â¯62) versus 140 mg/m2 (Mel-140, nâ¯=â¯283) in combination with Flu. Median age at allo-HCT was 66 years and median follow-up was 36.5 months. For Mel-100 versus Mel-140 groups, any grade gastrointestinal (GI) toxicity rates were 40.3% versus 67.8% (P < .001), day 100 grade II to IV acute graft-versus-host disease (GVHD) rates were 21.0% versus 43.1% (Pâ¯=â¯.001) and 2-year chronic GVHD rates were 17.4% versus 27.1% (Pâ¯=â¯.033). In multivariable analysis, Mel-140 resulted in higher risks of GI toxicity (HRâ¯=â¯1.83, Pâ¯=â¯.013), grade II to IV acute GVHD (HR=2.35, Pâ¯=â¯.003), and moderate/severe chronic GVHD (HRâ¯=â¯3.13, Pâ¯=â¯.007). Total Mel dose had no independent impact on oral mucositis, nonrelapse mortality, relapse, relapse-free survival, and overall survival. While independent validation of our observation is warranted, our findings support using Mel-100 in combination with Flu to minimize allo-HCT toxicities and morbidities related to GVHD.
RESUMO
Identifying novel and reliable prognostic biomarkers for predicting patient survival outcomes is essential for deciding personalized treatment strategies for diseases such as cancer. Numerous feature selection techniques have been proposed to address the high-dimensional problem in constructing prediction models. Not only does feature selection lower the data dimension, but it also improves the prediction accuracy of the resulted models by mitigating overfitting. The performances of these feature selection methods when applied to survival models, on the other hand, deserve further investigation. In this paper, we construct and compare a series of prediction-oriented biomarker selection frameworks by leveraging recent machine learning algorithms, including random survival forests, extreme gradient boosting, light gradient boosting and deep learning-based survival models. Additionally, we adapt the recently proposed prediction-oriented marker selection (PROMISE) to a survival model (PROMISE-Cox) as a benchmark approach. Our simulation studies indicate that boosting-based approaches tend to provide superior accuracy with better true positive rate and false positive rate in more complicated scenarios. For demonstration purpose, we applied the proposed biomarker selection strategies to identify prognostic biomarkers in different modalities of head and neck cancer data.
RESUMO
BACKGROUND: Alzheimer's disease (AD) is a chronic condition that progresses over time. While several therapeutic approaches have been developed, none have substantially altered disease progression. One explanation is that the disease is multi factorial. OBJECTIVE: Using the Affirmativ Health Personal Therapeutic Program (PTPr), we sought to determine whether a comprehensive and personalized program could improve cognitive and metabolic function in individuals diagnosed with subjective cognitive impairment, mild cognitive impairment, and early stage AD. METHODS: 35 individuals submitted blood samples and Montreal Cognitive Assessment (MoCA) scores, and answered intake questions. Individuals and caregivers participated in a four-day immersion program, which included Personal Therapeutic Plans (PTP), consultations with clinical practitioners, and explanations of the PTPr and PTP. Participants had follow-up by telemonitoring, with repeat blood sample analysis, updates regarding lifestyle choices, current medications and supplements, and MoCA testing at least once between 3 and 12 months after the PTPr. RESULTS: By comparing baseline to follow-up testing, we determined several risk factor scores, including blood glucose and insulin levels, and levels of vitamins B12, D3, and E, improved either in the entire participant pool or specifically in individuals with measures outside the normal range for specific factors. MoCA scores were stabilized in the entire participant pool and significantly improved in individuals scoring 24 or less at baseline. CONCLUSION: Our findings provide evidence that a comprehensive and personalized approach designed to mitigate AD risk factors can improve risk factor scores and stabilize cognitive function, warranting more extensive and placebo-controlled clinical studies.
RESUMO
BACKGROUND: MUC16 is a mucin marker that is frequently mutated in melanoma, but whether MUC16 mutations could be useful as a surrogate biomarker for tumor mutation burden (TMB) remains unclear. METHODS: This study rigorously evaluates the MUC16 mutation as a clinical biomarker in cutaneous melanoma by utilizing genomic and clinical data from patient samples from The Cancer Genome Atlas (TCGA) and two independent validation cohorts. We further extended the analysis to studies with patients treated with immunotherapies. RESULTS: Analysis results showed that samples with MUC16 mutations had a higher TMB than the samples of wild-type, with strong statistical significance (P < 0.001) in all melanoma cohorts tested. Associations between MUC16 mutations and TMB remained statistically significant after adjusting for potential confounding factors in the TCGA cohort [OR, 9.28 (95% confidence interval (CI), 5.18-17.39); P < 0.001], Moffitt cohort [OR, 31.95 (95% CI, 8.71-163.90); P < 0.001], and Yale cohort [OR, 8.09 (95% CI, 3.12-23.79); P < 0.01]. MUC16 mutations were also found to be associated with overall survival in the TCGA [HR, 0.62; (95% CI, 0.45-0.85); P < 0.01] and Moffitt cohorts [HR, 0.49 (95% CI, 0.28-0.87); P = 0.014]. Strikingly, MUC16 is the only top frequently mutated gene for which prognostic significance was observed. MUC16 mutations were also found valuable in predicting anti-CTLA-4 and anti-PD-1 therapy responses. CONCLUSIONS: MUC16 mutation appears to be a useful predictive marker of global TMB and patient survival in melanoma. IMPACT: This is, to the best of our knowledge, the first systematic evaluation of MUC16 mutation as a clinical biomarker and a predictive biomarker for immunotherapy in melanoma.