Comparison of the ABC and ACMG systems for variant classification.

The ABC and ACMG variant classification systems were compared by asking mainly European clinical laboratories to classify variants in 10 challenging cases using both systems, and to state if the variant in question would be reported as a relevant result or not as a measure of clinical utility. In contrast to the ABC system, the ACMG system was not made to guide variant reporting but to determine the likelihood of pathogenicity. Nevertheless, this comparison is justified since the ACMG class determines variant reporting in many laboratories. Forty-three laboratories participated in the survey. In seven cases, the classification system used did not influence the reporting likelihood when variants labeled as "maybe report" after ACMG-based classification were included. In three cases of population frequent but disease-associated variants, there was a difference in favor of reporting after ABC classification. A possible reason is that ABC step C (standard variant comments) allows a variant to be reported in one clinical setting but not another, e.g., based on Bayesian-based likelihood calculation of clinical relevance. Finally, the selection of ACMG criteria was compared between 36 laboratories. When excluding criteria used by less than four laboratories (<10%), the average concordance rate was 46%. Taken together, ABC-based classification is more clear-cut than ACMG-based classification since molecular and clinical information is handled separately, and variant reporting can be adapted to the clinical question and phenotype. Furthermore, variants do not get a clinically inappropriate label, like pathogenic when not pathogenic in a clinical context, or variant of unknown significance when the significance is known.

Do machine learning methods lead to similar individualized treatment rules? A comparison study on real data.

Identifying patients who benefit from a treatment is a key aspect of personalized medicine, which allows the development of individualized treatment rules (ITRs). Many machine learning methods have been proposed to create such rules. However, to what extent the methods lead to similar ITRs, that is, recommending the same treatment for the same individuals is unclear. In this work, we compared 22 of the most common approaches in two randomized control trials. Two classes of methods can be distinguished. The first class of methods relies on predicting individualized treatment effects from which an ITR is derived by recommending the treatment evaluated to the individuals with a predicted benefit. In the second class, methods directly estimate the ITR without estimating individualized treatment effects. For each trial, the performance of ITRs was assessed by various metrics, and the pairwise agreement between all ITRs was also calculated. Results showed that the ITRs obtained via the different methods generally had considerable disagreements regarding the patients to be treated. A better concordance was found among akin methods. Overall, when evaluating the performance of ITRs in a validation sample, all methods produced ITRs with limited performance, suggesting a high potential for optimism. For non-parametric methods, this optimism was likely due to overfitting. The different methods do not lead to similar ITRs and are therefore not interchangeable. The choice of the method strongly influences for which patients a certain treatment is recommended, drawing some concerns about their practical use.

Personalizing renal replacement therapy initiation in the intensive care unit: a reinforcement learning-based strategy with external validation on the AKIKI randomized controlled trials.

OBJECTIVE: The timely initiation of renal replacement therapy (RRT) for acute kidney injury (AKI) requires sequential decision-making tailored to individuals' evolving characteristics. To learn and validate optimal strategies for RRT initiation, we used reinforcement learning on clinical data from routine care and randomized controlled trials. MATERIALS AND METHODS: We used the MIMIC-III database for development and AKIKI trials for validation. Participants were adult ICU patients with severe AKI receiving mechanical ventilation or catecholamine infusion. We used a doubly robust estimator to learn when to start RRT after the occurrence of severe AKI for three days in a row. We developed a "crude strategy" maximizing the population-level hospital-free days at day 60 (HFD60) and a "stringent strategy" recommending RRT when there is significant evidence of benefit for an individual. For validation, we evaluated the causal effects of implementing our learned strategies versus following current best practices on HFD60. RESULTS: We included 3748 patients in the development set and 1068 in the validation set. Through external validation, the crude and stringent strategies yielded an average difference of 13.7 [95% CI -5.3 to 35.7] and 14.9 [95% CI -3.2 to 39.2] HFD60, respectively, compared to current best practices. The stringent strategy led to initiating RRT within 3 days in 14% of patients versus 38% under best practices. DISCUSSION: Implementing our strategies could improve the average number of days that ICU patients spend alive and outside the hospital while sparing RRT for many. CONCLUSION: We developed and validated a practical and interpretable dynamic decision support system for RRT initiation in the ICU.