Time-Dependent ROC Curve Analysis for Assessing the Capability of Radiation-Induced CD8 T-Lymphocyte Apoptosis to Predict Late Toxicities after Adjuvant Radiotherapy of Breast Cancer Patients.

Late fibrosis can occur in breast cancer patients treated with curative-intent radiotherapy. Predicting this toxicity is of clinical interest in order to adapt the irradiation dose delivered. Radiation-induced CD8 T-lymphocyte apoptosis (RILA) had been proven to be associated with less grade ≥2 late radiation-induced toxicities in patients with miscellaneous cancers. Tobacco smoking status and adjuvant hormonotherapy were also identified as potential factors related to late-breast-fibrosis-free survival. This article evaluates the predictive performance of the RILA using a ROC curve analysis that takes into account the dynamic nature of fibrosis occurrence. This time-dependent ROC curve approach is also applied to evaluate the ability of the RILA combined with the other previously identified factors. Our analysis includes a Monte Carlo cross-validation procedure and the calculation of an expected cost of misclassification, which provides more importance to patients who have no risk of late fibrosis in order to be able to treat them with the maximal irradiation dose. Performance evaluation was assessed at 12, 24, 36 and 50 months. At 36 months, our results were comparable to those obtained in a previous study, thus underlying the predictive power of the RILA. Based on specificity and cost, RILA alone seemed to be the most performant, while its association with the other factors had better negative predictive value results.

Flexible modeling of longitudinal health-related quality of life data accounting for informative dropout in a cancer clinical trial.

PURPOSE: A joint modeling approach is recommended for analysis of longitudinal health-related quality of life (HRQoL) data in the presence of potentially informative dropouts. However, the linear mixed model modeling the longitudinal HRQoL outcome in a joint model often assumes a linear trajectory over time, an oversimplification that can lead to incorrect results. Our aim was to demonstrate that a more flexible model gives more reliable and complete results without complicating their interpretation. METHODS: Five dimensions of HRQoL in patients with esophageal cancer from the randomized clinical trial PRODIGE 5/ACCORD 17 were analyzed. Joint models assuming linear or spline-based HRQoL trajectories were applied and compared in terms of interpretation of results, graphical representation, and goodness of fit. RESULTS: Spline-based models allowed arm-by-time interaction effects to be highlighted and led to a more precise and consistent representation of the HRQoL over time; this was supported by the martingale residuals and the Akaike information criterion. CONCLUSION: Linear relationships between continuous outcomes (such as HRQoL scores) and time are usually the default choice. However, the functional form turns out to be important by affecting both the validity of the model and the statistical significance. TRIAL REGISTRATION: This study is registered with ClinicalTrials.gov, number NCT00861094.

Should we use liver grafts repeatedly refused by other transplant teams?

BACKGROUND & AIMS: In France, liver grafts that have been refused at least 5 times can be "rescued" and allocated to a centre which chooses a recipient from its own waiting list, outside the patient-based allocation framework. We explored whether these "rescued" grafts were associated with worse graft/patient survival, as well as assessing their effect on survival benefit. METHODS: Among 7,895 candidates, 5,218 were transplanted between 2009 and 2014 (336 centre-allocated). We compared recipient/graft survival between patient allocation and centre allocation, considering a selection bias and the distribution of centre-allocation recipients among the transplant teams. We used a propensity score approach and a weighted Cox model using the inverse probability of treatment weighting method. We also explored the survival benefit associated with centre-allocation grafts. RESULTS: There was a significantly higher risk of graft loss/death in the centre allocation group compared to the patient allocation group (hazard ratio 1.13; 95% CI 1.05-1.22). However, this difference was no longer significant for teams that performed more than 7% of the centre-allocation transplantations. Moreover, receiving a centre-allocation graft, compared to remaining on the waiting list and possibly later receiving a patient-allocation graft, did not convey a poorer survival benefit (hazard ratio 0.80; 95% CI 0.60-1.08). CONCLUSIONS: In centres which transplanted most of the centre-allocation grafts, using grafts repeatedly refused for top-listed candidates was not detrimental. Given the organ shortage, our findings should encourage policy makers to restrict centre-allocation grafts to targeted centres. LAY SUMMARY: "Centre allocation" (CA) made it possible to save 6 out of 100 available liver grafts that had been refused at least 5 times for use in the top-listed candidates on the national waiting list. In this series, the largest on this topic, we showed that, in centres which transplanted most of the CA grafts, using grafts repeatedly refused for top-listed candidates did not appear to be detrimental. In the context of organ shortage, our results, which could be of interest for any country using this CA strategy, should encourage policy makers to reassess some aspects of graft allocation by restricting CA grafts to targeted centres, fostering the "best" matching between grafts and candidates on the waiting list.

Matching Graft Quality to Recipient's Disease Severity Based on the Survival Benefit in Liver Transplantation.

Persistent shortage and heterogeneous quality of liver grafts encourages the optimization of donor-recipient matching in liver transplantation (LT). We explored whether or not there was a survival benefit (SB) of LT according to the quality of grafts assessed by the Donor Quality Index (DQI) and recipients' disease severity, using the Model for End-Stage Liver Disease (MELD) in 8387 French patients wait-listed between 2009 and 2014. SB associated with LT was estimated using the sequential stratification method in different categories of MELD and DQI. For each transplantation, a stratum was created that matched one transplanted patient with all eligible control candidates. Strata were thereafter combined, and a stratified Cox model, adjusted for covariates, was fitted in order to estimate hazard ratios that qualified the SB according to each MELD and DQI sub-group. A significant SB was observed for all MELD and DQI sub-groups, with the exception of high MELD patients transplanted with "high-risk" grafts. More specifically, in decompensated-cirrhosis patients, "high-risk" grafts did not appear to be detrimental in medium MELD patients. Interestingly, in hepatocellular-carcinoma (HCC) patients, a significant SB was found for all MELD-DQI combinations. For MELD exceptions no SB was found. In terms of SB, "low-risk" grafts appeared appropriate for most severe patients (MELD > 30). Conversely, low/medium MELD and HCC patients presented an SB while allocated "high-risk" grafts. Thus, SB based matching rules for LT candidates might improve the survival of the LT population as a whole.

Liver Transplantation with "Hors Tour" Allocated Versus Standard MELD Allocated Grafts: Single-Center Audit and Impact on the Liver Pool in France.

BACKGROUND: The French transplant governing system defined "Rescue" (the so-called "Hors Tour") livers as those livers which were declined for the five top-listed patients. This study compares the outcomes following liver transplantation (LT) in patients who received a donor liver through a rescue allocation (RA) procedure or according to MELD score priority (standard allocation, SA) and evaluates the impact on the graft pool of a proactive policy to accept RA grafts. METHODS: Data from all consecutive patients who underwent LT with SA or RA grafts from 2011 to 2015 were compared in terms of short- and long-term outcomes. RESULTS: The 249 elective first LTs were performed with 64 (25.7%) RA and 185 (74.3%) SA grafts. RA grafts were obtained from older donors and were associated with a longer cold ischemia time. Recipients of RA livers were older and had lower MELD scores. The rates of delayed graft function, primary nonfunction, retransplantation, complications, and mortality were similar between the RA and SA groups. At 1 and 3 and 5 years, graft and patient survival rates were similar between the groups. These results were maintained after matching on recipient characteristics. Our proactive policy to accept RA grafts increased the liver pool for elective first transplantation by 25%. CONCLUSIONS: RA livers can be safely transplanted into selected recipients and significantly expand the liver pool.

External validation and recalibration of the Brock model to predict probability of cancer in pulmonary nodules using NLST data.

INTRODUCTION: We performed an external validation of the Brock model using the National Lung Screening Trial (NLST) data set, following strict guidelines set forth by the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis statement. We report how external validation results can be interpreted and highlight the role of recalibration and model updating. MATERIALS AND METHODS: We assessed model discrimination and calibration using the NLST data set. Adhering to the inclusion/exclusion criteria reported by McWilliams et al, we identified 7879 non-calcified nodules discovered at the baseline low-dose CT screen with 2 years of follow-up. We characterised differences between Pan-Canadian Early Detection of Lung Cancer Study and NLST cohorts. We calculated the slope on the prognostic index and the intercept coefficient by fitting the original Brock model to NLST. We also assessed the impact of model recalibration and the addition of new covariates such as body mass index, smoking status, pack-years and asbestos. RESULTS: While the area under the curve (AUC) of the model was good, 0.905 (95% CI 0.882 to 0.928), a histogram plot showed that the model poorly differentiated between benign and malignant cases. The calibration plot showed that the model overestimated the probability of cancer. In recalibrating the model, the coefficients for emphysema, spiculation and nodule count were updated. The updated model had an improved calibration and achieved an optimism-corrected AUC of 0.912 (95% CI 0.891 to 0.932). Only pack-year history was found to be significant (p<0.01) among the new covariates evaluated. CONCLUSION: While the Brock model achieved a high AUC when validated on the NLST data set, the model benefited from updating and recalibration. Nevertheless, covariates used in the model appear to be insufficient to adequately discriminate malignant cases.

Using Sequential Decision Making to Improve Lung Cancer Screening Performance.

Globally, lung cancer is responsible for nearly one in five cancer deaths. The National Lung Screening Trial (NLST) demonstrated the efficacy of low-dose computed tomography (LDCT) to identify early-stage disease, setting the basis for widespread implementation of lung cancer screening programs. However, the specificity of LDCT lung cancer screening is suboptimal, with a significant false positive rate. Representing this imaging-based screening process as a sequential decision making problem, we combined multiple machine learning-based methods to learn a partially-observable Markov decision process that simultaneously optimizes lung cancer detection while enhancing test specificity. Using NLST data, we trained a dynamic Bayesian network as an observational model and used inverse reinforcement learning to discover a rewards function based on experts' decisions. Our resultant predictive model decreased the false positive rate while maintaining a high true positive rate at a level comparable to human experts. Our model also detected a number of lung cancers earlier.