From algorithms to action: improving patient care requires causality.

In cancer research there is much interest in building and validating outcome prediction models to support treatment decisions. However, because most outcome prediction models are developed and validated without regard to the causal aspects of treatment decision making, many published outcome prediction models may cause harm when used for decision making, despite being found accurate in validation studies. Guidelines on prediction model validation and the checklist for risk model endorsement by the American Joint Committee on Cancer do not protect against prediction models that are accurate during development and validation but harmful when used for decision making. We explain why this is the case and how to build and validate models that are useful for decision making.

Robustness of pulmonary nodule radiomic features on computed tomography as a function of varying radiation dose levels-a multi-dose in vivo patient study.

OBJECTIVE: Analysis of textural features of pulmonary nodules in chest CT, also known as radiomics, has several potential clinical applications, such as diagnosis, prognostication, and treatment response monitoring. For clinical use, it is essential that these features provide robust measurements. Studies with phantoms and simulated lower dose levels have demonstrated that radiomic features can vary with different radiation dose levels. This study presents an in vivo stability analysis of radiomic features for pulmonary nodules against varying radiation dose levels. METHODS: Nineteen patients with a total of thirty-five pulmonary nodules underwent four chest CT scans at different radiation dose levels (60, 33, 24, and 15 mAs) in a single session. The nodules were manually delineated. To assess the robustness of features, we calculated the intra-class correlation coefficient (ICC). To visualize the effect of milliampere-second variation on groups of features, a linear model was fitted to each feature. We calculated bias and calculated the R2 value as a measure of goodness of fit. RESULTS: A small minority of 15/100 (15%) radiomic features were considered stable (ICC > 0.9). Bias increased and R2 decreased at lower dose, but shape features seemed to be more robust to milliampere-second variations than other feature classes. CONCLUSION: A large majority of pulmonary nodule radiomic features were not inherently robust to radiation dose level variations. For a subset of features, it was possible to correct this variability by a simple linear model. However, the correction became increasingly less accurate at lower radiation dose levels. CLINICAL RELEVANCE STATEMENT: Radiomic features provide a quantitative description of a tumor based on medical imaging such as computed tomography (CT). These features are potentially useful in several clinical tasks such as diagnosis, prognosis prediction, treatment effect monitoring, and treatment effect estimation. KEY POINTS: • The vast majority of commonly used radiomic features are strongly influenced by variations in radiation dose level. • A small minority of radiomic features, notably the shape feature class, are robust against dose-level variations according to ICC calculations. • A large subset of radiomic features can be corrected by a linear model taking into account only the radiation dose level.

The Association between Muscle Quantity and Overall Survival Depends on Muscle Radiodensity: A Cohort Study in Non-Small-Cell Lung Cancer Patients.

The prognostic value of CT-derived muscle quantity for overall survival (OS) in patients with non-small-cell lung cancer (NSCLC) is uncertain due to conflicting evidence. We hypothesize that increased muscle quantity is associated with better OS in patients with normal muscle radiodensity but not in patients with fatty degeneration of muscle tissue and low muscle radiodensity. We performed an observational cohort study in NSCLC patients treated with radiotherapy. A deep learning algorithm was used to measure muscle quantity as psoas muscle index (PMI) and psoas muscle radiodensity (PMD) on computed tomography. The potential interaction between PMI and PMD for OS was investigated using Cox proportional-hazards regression. Baseline adjustment variables were age, sex, histology, performance score and body mass index. We investigated non-linear effects of continuous variables and imputed missing values using multiple imputation. We included 2840 patients and observed 1975 deaths in 5903 patient years. The average age was 68.9 years (standard deviation 10.4, range 32 to 96) and 1692 patients (59.6%) were male. PMI was more positively associated with OS for higher values of PMD (hazard ratio for interaction 0.915; 95% confidence interval 0.861-0.972; p-value 0.004). We found evidence that high muscle quantity is associated with better OS when muscle radiodensity is higher, in a large cohort of NSCLC patients treated with radiotherapy. Future studies on the association between muscle status and OS should accommodate this interaction in their analysis for more accurate and more generalizable results.

Individual treatment effect estimation in the presence of unobserved confounding using proxies: a cohort study in stage III non-small cell lung cancer.

Randomized Controlled Trials (RCT) are the gold standard for estimating treatment effects but some important situations in cancer care require treatment effect estimates from observational data. We developed "Proxy based individual treatment effect modeling in cancer" (PROTECT) to estimate treatment effects from observational data when there are unobserved confounders, but proxy measurements of these confounders exist. We identified an unobserved confounder in observational cancer research: overall fitness. Proxy measurements of overall fitness exist like performance score, but the fitness as observed by the treating physician is unavailable for research. PROTECT reconstructs the distribution of the unobserved confounder based on these proxy measurements to estimate the treatment effect. PROTECT was applied to an observational cohort of 504 stage III non-small cell lung cancer (NSCLC) patients, treated with concurrent chemoradiation or sequential chemoradiation. Whereas conventional confounding adjustment methods seemed to overestimate the treatment effect, PROTECT provided credible treatment effect estimates.

Prognostic factors for overall survival of stage III non-small cell lung cancer patients on computed tomography: A systematic review and meta-analysis.

INTRODUCTION: Prognosis prediction is central in treatment decision making and quality of life for non-small cell lung cancer (NSCLC) patients. However, conventional computed tomography (CT) related prognostic factors may not apply to the challenging stage III NSCLC group. The aim of this systematic review was therefore to identify and evaluate CT-related prognostic factors for overall survival (OS) of stage III NSCLC. METHODS: The Medline, Embase, and Cochrane electronic databases were searched. After study selection, risk of bias was estimated for the included studies. Meta-analysis of univariate results was performed when sufficient data were available. RESULTS: 1595 of the 11,996 retrieved records were selected for full text review, leading to inclusion of 65 studies that reported data of 144,513 stage III NSCLC patients andcompromising 26 unique CT-related prognostic factors. Relevance and validity varied substantially, few studies had low relevance and validity. Only four studies evaluated the added value of new prognostic factors compared with recognized clinical factors. Included studies suggested gross tumor volume (meta-analysis: HR = 1.22, 95%CI: 1.05-1.42), tumor diameter, nodal volume, and pleural effusion, are prognostic in patients treated with chemoradiation. Clinical T-stage and location (right/left) were likely not prognostic within stage III NSCLC. Inconclusive are several radiomic features, tumor volume, atelectasis, location (pulmonary lobes, central/peripheral), interstitial lung abnormalities, great vessel invasion, pit-fall sign, and cavitation. CONCLUSIONS: Tumor-size and nodal size-related factors are prognostic for OS in stage III NSCLC. Future studies should carefully report study characteristics and contrast factors with guideline recognized factors to improve evidence evaluation and validation.

First Evidence for a Dose-Response Relationship in Patients Treated with 166Ho Radioembolization: A Prospective Study.

166Ho-microspheres have recently been approved for clinical use for hepatic radioembolization in the European Union. The aim of this study was to investigate the absorbed dose-response relationship and its association with overall survival for 166Ho radioembolization in patients with liver metastases. Methods: Patients treated in the HEPAR I and II studies who underwent an 18F-FDG PET/CT scan at baseline, a posttreatment 166Ho SPECT/CT scan, and another 18F-FDG PET/CT scan at the 3-mo follow-up were included for analysis. The posttreatment 166Ho-microsphere activity distributions were estimated with quantitative SPECT/CT reconstructions using a quantitative Monte Carlo-based method. The response of each tumor was based on the change in total lesion glycolysis (TLG) between baseline and follow-up and was placed into 1 of 4 categories, according to the PERCIST criteria, ranging from complete response to progressive disease. Patient-level response was grouped according to the average change in TLG per patient. The absorbed dose-response relationship was assessed using a linear mixed model to account for correlation of tumors within patients. Median overall survival was compared between patients with and without a metabolic liver response, using a log-rank test. Results: Thirty-six patients with a total of 98 tumors were included. The relation between tumor-absorbed dose and both tumor-level and patient-level response was explored. At a tumor level, a significant difference in geometric mean absorbed dose was found between complete response (232 Gy; 95% confidence interval [CI], 178-303 Gy; n = 32) and stable disease (147 Gy; 95% CI, 113-191 Gy; n = 28) (P = 0.01) and between complete response and progressive disease (117 Gy; 95% CI, 87-159 Gy; n = 21) (P = 0.0008). This constitutes a robust absorbed dose-response relationship. At a patient level, a significant difference was found between patients with complete or partial response (210 Gy; 95% CI, 161-274 Gy; n = 13) and patients with progressive disease (116 Gy; 95% CI, 81-165 Gy; n = 9) (P = 0.01). Patients were subsequently grouped according to their average change in TLG. Patients with an objective response (complete or partial) exhibited a significantly higher overall survival than nonresponding patients (stable or progressive disease) (median, 19 mo vs. 7.5 mo; log-rank, P = 0.01). Conclusion: These results confirm a significant absorbed dose-response relationship in 166Ho radioembolization. Treatment response is associated with a higher overall survival.

The morphological substrate for Renal Denervation: Nerve distribution patterns and parasympathetic nerves. A post-mortem histological study.

BACKGROUND: Renal Denervation as a possible treatment for hypertension has been studied extensively, but knowledge on the distribution of nerves surrounding the renal artery is still incomplete. While sympathetic and sensory nerves have been demonstrated, there is no mention of the presence of parasympathetic nerve fibers. OBJECTIVES: To provide a description of the distribution patterns of the renal nerves in man, and, in addition, provide a detailed representation of the relative contribution of the sympathetic, parasympathetic and afferent divisions of the autonomic nervous system. METHODS: Renal arteries of human cadavers were each divided into four longitudinal segments and immunohistochemically stained with specific markers for afferent, parasympathetic and sympathetic nerves. Nerve fibers were semi-automatically quantified by computerized image analysis, and expressed as cross-sectional area relative to the distance to the lumen. RESULTS: A total of 3372 nerve segments were identified in 8 arteries of 7 cadavers. Sympathetic, parasympathetic and afferent nerves contributed for 73.5% (95% CI: 65.4-81.5%), 17.9% (10.7-25.1%) and 8.7% (5.0-12.3%) of the total cross-sectional nerve area, respectively. Nerves are closer to the lumen in more distal segments and larger bundles that presumably innervate the kidney lie at 1-3.5mm distance from the lumen. The tissue-penetration depth of the ablation required to destroy 50% of the nerve fibers is 2.37 mm in the proximal segment and 1.78 mm in the most distal segments. CONCLUSION: Sympathetic, parasympathetic and afferent nerves exist in the vicinity of the renal artery. The results warrant further investigation of the role of the parasympathetic nervous system on renal physiology, and may contribute to refinement of the procedure by focusing the ablation on the most distal segment.