Search

The proposal of improving reproducibility by lowering the significance threshold to 0.005 has been discussed, but the impact on conducting clinical trials has yet to be examined from a study design perspective. The impact on sample size and study duration was investigated using design setups from 125 phase II studies published between 2015 and 2022. The impact was assessed using percent increase in sample size and additional years of accrual with the medians being 110.97% higher and 2.65 years longer respectively. The results indicated that this proposal causes additional financial burdens that reduce the efficiency of conducting clinical trials.

Abstract Background Amiodarone is one of the most commonly used anti-arrhythmic agents. Amiodarone pulmonary toxicity is a potentially fatal adverse effect associated with amiodarone use. Previous studies on the epidemiology and risk factors for amiodarone pulmonary toxicity showed diverse results. Methods A multicenter retrospective cohort study was conducted to identify clinic-epidemiologic markers associated with amiodarone pulmonary toxicity for development of a prediction rule. Patients taking amiodarone who were managed in 3 centres in Hong Kong from 2005 to 2015 were included in this study. Penalized logistic regression was used to model the outcome as it is rare. Results A total of 34 cases with amiodarone pulmonary toxicity were identified among 1786 patients taking amiodarone for at least 90 days from 2005 to 2015. The incidence of amiodarone pulmonary toxicity was estimated to be 1.9%. The risk factors for amiodarone pulmonary toxicity included advanced age (OR 1.047, 95% CI 1.010–1.085, p = 0.013), ventricular arrhythmia (OR 2.703, 95% CI 1.053–6.935, p = 0.039), underlying lung disease (OR 2.511, 95% CI 1.146–5.501, p = 0.021) and cumulative dose of amiodarone (OR 4.762, 95% CI 1.310–17.309 p = 0.018). Conclusions The incidence of amiodarone pulmonary toxicity in Chinese patients in Hong Kong is estimated to be 1.9% in this study. Age, underlying lung disease, ventricular arrhythmia and cumulative dose of amiodarone are associated with the development of amiodarone pulmonary toxicity. A prediction rule was developed to inform the risk of developing amiodarone pulmonary toxicity.

There has been a growing trend that activities relating to clinical trials take place at locations other than traditional trial sites (hence decentralized clinical trials or DCTs), some of which are at settings of real-world clinical practice. Although there are numerous benefits of DCTs, this also brings some implications on a number of issues relating to the design, conduct, and analysis of DCTs. The Real-World Evidence Scientific Working Group of the American Statistical Association Biopharmaceutical Section has been reviewing the field of DCTs and provides in this paper considerations for decentralized trials from a statistical perspective. This paper first discusses selected critical decentralized elements that may have statistical implications on the trial and then summarizes regulatory guidance, framework, and initiatives on DCTs. More discussions are presented by focusing on the design (including construction of estimand), implementation, statistical analysis plan (including missing data handling), and reporting of safety events. Some additional considerations (e.g., ethical considerations, technology infrastructure, study oversight, data security and privacy, and regulatory compliance) are also briefly discussed. This paper is intended to provide statistical considerations for decentralized trials of medical products to support regulatory decision-making., (© 2025 Novartis Pharmaceuticals Corporation. Pfizer Inc. AbbVie Inc and The Author(s). Clinical and Translational Science published by Wiley Periodicals LLC on behalf of American Society for Clinical Pharmacology and Therapeutics. This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.)

Whereas AI/ML methods were considered experimental tools in clinical development for some time, nowadays they are widely available. However, stakeholders in the health care industry still need to answer the question which role these methods can realistically play and what standards should be adhered to. Clinical research in late-stage clinical development has particular requirements in terms of robustness, transparency and traceability. These standards should also be adhered to when applying AI/ML methods. Currently there is some formal regulatory guidance available, but this is more directed at settings where a device or medical software is investigated. Here we focus on the application of AI/ML methods in late-stage clinical drug development, i.e. in a setting where currently less guidance is available. This is done via first summarizing available regulatory guidance and work done by regulatory statisticians followed by the presentation of an industry application where the influence of extensive sets of baseline characteristics on the treatment effect can be investigated by applying ML-methods in a standardized manner with intuitive graphical displays leveraging explainable AI methods. The paper aims at stimulating discussions on the role such analyses can play in general rather than advocating for a particular AI/ML-method or indication where such methods could be meaningful., (© 2024. The Author(s), under exclusive licence to The Drug Information Association, Inc.)

Background: The use of historical external control data in clinical trials has grown in interest and needs when considering the design of future trials. Hybrid control designs can be more efficient to achieve the same power with fewer patients and limited resources. The literature is sparse on appropriate statistical methods which can account for the differences between historical external controls and the control patients in a study. In this article, we illustrate the analysis framework of a clinical trial if a hybrid control design was used after determining an RCT may not be feasible., Methods: We utilize two previously completed RCTs in nonsquamous NSCLC and a nationwide electronic health record derived de-identified database as examples and compare 5 analysis methods on each trial, as well as a set of simulations to determine operating characteristics of such designs., Results: In single trial estimation, the Case Weighted Adaptive Power Prior provided estimated treatment hazard ratios consistent with the original trial's conclusions with narrower confidence intervals. The simulation studies showed that the Case Weighted Adaptive Power Prior achieved the highest power (and well controlled type-1 error) across all 5 methods with consistent study sample size., Conclusions: By following the proposed hybrid control framework, one can design a hybrid control trial transparently and accounting for differences between control groups while controlling type-1 error and still achieving efficiency gains from the additional contribution from external controls., (© 2024. The Author(s).)

We consider the challenges associated with causal inference in settings where data from a randomized trial are augmented with control data from an external source to improve efficiency in estimating the average treatment effect (ATE). This question is motivated by the SUNFISH trial, which investigated the effect of risdiplam on motor function in patients with spinal muscular atrophy. While the original analysis used only data generated by the trial, we explore an alternative analysis incorporating external controls from the placebo arm of a historical trial. We cast the setting into a formal causal inference framework and show how these designs are characterized by a lack of full randomization to treatment and heightened dependency on modeling. To address this, we outline sufficient causal assumptions about the exchangeability between the internal and external controls to identify the ATE and establish a connection with novel graphical criteria. Furthermore, we propose estimators, review efficiency bounds, develop an approach for efficient doubly robust estimation even when unknown nuisance models are estimated with flexible machine learning methods, suggest model diagnostics, and demonstrate finite-sample performance of the methods through a simulation study. The ideas and methods are illustrated through their application to the SUNFISH trial, where we find that external controls can increase the efficiency of treatment effect estimation., (© The Author(s) 2024. Published by Oxford University Press on behalf of The International Biometric Society.)

In clinical trials, it is common to design a study that permits the administration of an experimental treatment to participants in the placebo or standard of care group post primary endpoint. This is often seen in the open-label extension phase of a phase III, pivotal study of the new medicine, where the focus is on assessing long-term safety and efficacy. With the availability of external controls, proper estimation and inference of long-term treatment effect during the open-label extension phase in the absence of placebo-controlled patients are now feasible. Within the framework of causal inference, we propose several difference-in-differences (DID) type methods and a synthetic control method (SCM) for the combination of randomized controlled trials and external controls. Our realistic simulation studies demonstrate the desirable performance of the proposed estimators in a variety of practical scenarios. In particular, DID methods outperform SCM and are the recommended methods of choice. An empirical application of the methods is demonstrated through a phase III clinical trial in rare disease.

We develop a method for hybrid analyses that uses external controls to augment internal control arms in randomized controlled trials (RCTs) where the degree of borrowing is determined based on similarity between RCT and external control patients to account for systematic differences (e.g., unmeasured confounders). The method represents a novel extension of the power prior where discounting weights are computed separately for each external control based on compatibility with the randomized control data. The discounting weights are determined using the predictive distribution for the external controls derived via the posterior distribution for time-to-event parameters estimated from the RCT. This method is applied using a proportional hazards regression model with piecewise constant baseline hazard. A simulation study and a real-data example are presented based on a completed trial in non-small cell lung cancer. It is shown that the case weighted power prior provides robust inference under various forms of incompatibility between the external controls and RCT population., (© The Author(s) 2024. Published by Oxford University Press on behalf of The International Biometric Society.)

Phase Ib/II oncology trials, despite their small sample sizes, aim to provide information for optimal internal company decision-making concerning novel drug development. Hybrid controls (a combination of the current control arm and controls from one or more sources of historical trial data [HTD]) can be used to increase statistical precision. Here we assess combining two sources of Roche HTD to construct a hybrid control in targeted therapy for decision-making via an extensive simulation study. Our simulations are based on the real data of one of the experimental arms and the control arm of the MORPHEUS-UC Phase Ib/II study and two Roche HTD for atezolizumab monotherapy. We consider potential complications such as model misspecification, unmeasured confounding, different sample sizes of current treatment groups, and heterogeneity among the three trials. We evaluate two frequentist methods (with both Cox and Weibull accelerated failure time [AFT] models) and three different commensurate priors in Bayesian dynamic borrowing (with a Weibull AFT model), and modifications within each of those, when estimating the effect of treatment on survival outcomes and measures of effect such as marginal hazard ratios. We assess the performance of these methods in different settings and the potential of generalizations to supplement decisions in early-phase oncology trials. The results show that the proposed joint frequentist methods and noninformative priors within Bayesian dynamic borrowing with no adjustment on covariates are preferred, especially when treatment effects across the three trials are heterogeneous. For generalization of hybrid control methods in such settings, we recommend more simulation studies., (© 2023 John Wiley & Sons Ltd.)

Background: Despite the growth in primary cancer (PC) survivors, the trends and disparities in this population have yet to be comprehensively examined using competing risk analysis. The objective is to examine trends in time to second primary cancer (SPC) and to characterize age, sex, and racial disparities in time-to-SPC., Methods: A retrospective analysis was conducted based on Surveillance, Epidemiology, and End Results (SEER). Two datasets for this study are (1) the discovery dataset with patients from SEER-8 (1990-2019) and (2) the validation dataset with patients from SEER-17 (2000-2019), excluding those in the discovery dataset. Patients were survivors of lung, colorectal, breast (female only), and prostate PCs., Results: The 5-year SPC cumulative incidences of lung PC increased from 1990 to 2019, with the cumulative incidence ratio being 1.73 (95% confidence intervals [CI], 1.64-1.82; p < 0.001). Age disparities among all PCs remained from 2010 to 2019, and the adjusted HRs (aHRs) of all PCs were above 1.43 when those below 65 were compared with those 65 and above. Sex disparity exists among colorectal and lung PC survivors. Racial disparities existed among non-Hispanic (NH) Black breast PC survivors (aHR: 1.11; 95% CI: 1.07-1.17; p < 0.001). The types of SPC vary according to PC and sex., Conclusions: Over the past three decades, there has been a noticeably shortened time-to-SPC among lung PC survivors. This is likely attributed to the reduced number of lung cancer deaths due to advancements in effective treatments. However, disparities in age, sex, and race still exist, indicating that further effort is needed to close the gap., (© 2023 The Authors. Cancer Medicine published by John Wiley & Sons Ltd.)

Purpose: To investigate the enrollment success rate of cancer clinical trials conducted in 2008-2019 and various factors lowering the enrollment success rate., Methods: This is a cross-sectional study with clinical trial information from the largest registration database ClinicalTrials.gov. Enrollment success rate was defined as actual enrollment greater or equal to 85% of the estimated enrollment goal. The association between trial characteristics and enrollment success was evaluated using the multivariable logistic regression., Results: A total of 4,004 trials in breast, lung, and colorectal cancers were included. The overall enrollment success rate was 49.1%. Compared with 2008-2010 (51.5%) and 2011-2013 (52.1%), the enrollment success rate is lower in 2014-2016 (46.5%) and 2017-2019 (36.4%). Regression analyses found trial activation year, phase I, phase I/phase II, and phase II ( v phase III), sponsor agency of government ( v industry), not requiring healthy volunteers, and estimated enrollment of 50-100, 100-200, 200, and >500 ( v 0-50) were associated with a lower enrollment success rate ( P < .05). However, trials with placebo comparator, ≥5 locations ( v 1 location), and a higher number of secondary end points (eg, ≥5 v 0) were associated with a higher enrollment success rate ( P < .05). The AUC for prediction of the final logistic regression models for all trials and specific trial groups ranged from 0.69 to 0.76., Conclusion: This large-scale study supports a lower enrollment success rate over years in cancer clinical trials. Identified factors for enrollment success can be used to develop and improve recruitment strategies for future cancer trials.

Background and Objective: The application of immunotherapy in cancers, including liver cancer, has been increasing. However, non-proportional hazard (NPH) is often observed in cancer immunotherapy trials. In presence of violation of proportional hazard (PH) assumption, restricted mean survival time (RMST) ratio was proposed as an alternative to hazard ratio (HR) for evaluating the treatment effects of such trials. To shorten the total study duration, an intermediate endpoint with shorter follow-up such as progression-free survival (PFS) is used as the primary endpoint. Our aim is to evaluate the applicability of RMST ratio in addition to the HR in assessing the level of PFS serving as a surrogacy of overall survival (OS)., Methods: Phase II or phase III hepatocellular carcinoma (HCC) immunotherapy studies that were published between January 2013 and August 2022 were identified via the search in PubMed. Weighted least-square regression (WLSR) was applied to analyze the trial level data with the sample size of study being set as the weight. The evaluation was conducted twice with RMST ratio and HR being applied in respective evaluation to examine the level of PFS as a surrogacy for OS., Key Content and Findings: Based on the results of eight included trials, the R-square values of WLSR with either HR or RMST ratio being applied were 0.31 and 0.16 separately, indicating a moderate and low correlation between PFS and OS respectively., Conclusions: In this study, our results demonstrated the potential of RMST ratio in addition to HR for evaluating the level of surrogacy in immunotherapy trials. Furthermore, including more large scale and homogeneous studies into the research may help better understand the level of surrogacy in liver cancer.

According to the Global Burden of Disease Study, 3 million deaths were attributable to high sodium intake and low intake of whole grains. With the rapid evolution of the food industry, we hypothesize that dietary intake of sodium has increased and that dietary intake of whole grains or fibers has decreased because of easier access to highly processed food. Country-level data on dietary factors and country income levels from 1990 to 2018 were collected from 3 public databases. The trend of dietary intake was modeled using the linear mixed model accounting for random effects of individual countries. The country-level differences in dietary factors between males and females were calculated, and the trends were also modeled accounting for the random effects of countries. Both males and females consumed increasing amounts of dietary sodium from 1990 to 2018 in high-income, middle- to high-, middle-, and low-income countries. Dietary fiber intake increased in low-to-middle, middle-, and middle-to-high income countries for both men and women over the past 3 decades. Men tend to consume more sodium and less fiber and whole grains in their diets than women, the trend of which is statistically significant in middle-income countries. Over the past 3 decades, the macrotrend of dietary sodium has increased around the globe. To reduce the sodium intake level, nutrition policy should emphasize sodium reduction, especially in high-income, middle- to high-income, middle-income, and low-income countries., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Borrowing data from external control has been an appealing strategy for evidence synthesis when conducting randomized controlled trials (RCTs). Often named hybrid control trials, they leverage existing control data from clinical trials or potentially real-world data (RWD), enable trial designs to allocate more patients to the novel intervention arm, and improve the efficiency or lower the cost of the primary RCT. Several methods have been established and developed to borrow external control data, among which the propensity score methods and Bayesian dynamic borrowing framework play essential roles. Noticing the unique strengths of propensity score methods and Bayesian hierarchical models, we utilize both methods in a complementary manner to analyze hybrid control studies. In this article, we review methods including covariate adjustments, propensity score matching and weighting in combination with dynamic borrowing and compare the performance of these methods through comprehensive simulations. Different degrees of covariate imbalance and confounding are examined. Our findings suggested that the conventional covariate adjustment in combination with the Bayesian commensurate prior model provides the highest power with good type I error control under the investigated settings. It has desired performance especially under scenarios of different degrees of confounding. To estimate efficacy signals in the exploratory setting, the covariate adjustment method in combination with the Bayesian commensurate prior is recommended., (© 2023 The Authors. Pharmaceutical Statistics published by John Wiley & Sons Ltd.)

Background and Objective: Fully-supervised learning approaches have shown promising results in some health status prediction tasks using Electronic Health Records (EHRs). These traditional approaches rely on sufficient labeled data to learn from. However, in practice, acquiring large-scaled labeled medical data for various prediction tasks is often not feasible. Thus, it is of great interest to utilize contrastive pre-training to leverage the unlabeled information., Methods: In this work, we propose a novel data-efficient framework, contrastive predictive autoencoder (CPAE), to first learn without labels from the EHR data in the pre-training process, and then fine-tune on the downstream tasks. Our framework comprises of two parts: (i) a contrastive learning process, inherited from contrastive predictive coding (CPC), which aims to extract global slow-varying features, and (ii) a reconstruction process, which forces the encoder to capture local features. We also introduce the attention mechanism in one variant of our framework to balance the above two processes., Results: Experiments on real-world EHR dataset verify the effectiveness of our proposed framework on two downstream tasks (i.e., in-hospital mortality prediction and length-of-stay prediction), compared to their supervised counterparts, the CPC model, and other baseline models., Conclusions: By comprising of both contrastive learning components and reconstruction components, CPAE aims to extract both global slow-varying information and local transient information. The best results on two downstream tasks are all achieved by CPAE. The variant AtCPAE is particularly superior when fine-tuned on very small training data. Further work may incorporate techniques of multi-task learning to optimize the pre-training process of CPAEs. Moreover, this work is based on the benchmark MIMIC-III dataset which only includes 17 variables. Future work may extend to a larger number of variables., Competing Interests: Declaration of Competing Interest Authors declare that they have no conflict of interest., (Copyright © 2023 Elsevier B.V. All rights reserved.)

Bayesian historical borrowing has recently attracted growing interest due to the increasing availability of historical control data, as well as improved computational methodology and software. In this article, we argue that the statistical models used for borrowing may be suboptimal when they do not adjust for differing factors across historical studies such as covariates, dosing regimen, etc. We propose an alternative approach to address these shortcomings. We start by constructing a historical model based on subject-level historical data to accurately characterize the control treatment by adjusting for known between trials differences. This model is subsequently used to predict the control arm response in the current trial, enabling the derivation of a model-informed prior for the treatment effect parameter of another (potentially simpler) model used to analyze the trial efficacy (i.e. the trial model). Our approach is applied to neovascular age-related macular degeneration trials, employing a cross-sectional regression trial model, and a longitudinal non-linear mixed-effects drug-disease-trial historical model. The latter model characterizes the relationship between clinical response, drug exposure and baseline covariates so that the derived model-informed prior seamlessly adapts to the trial population and can be extrapolated to a different dosing regimen. This approach can yield a more accurate prior for borrowing, thus optimizing gains in efficiency (e.g. increasing power or reducing the sample size) in future trials.

Background: Appropriate analyses and reporting are essential to the reproducibility and interpretation of clinical trials. However, the coronavirus disease 19 (COVID-19) pandemic and other force majeure events, like the war in Ukraine, have impacted the conduct of clinical trials., Methods: The number of clinical trials potentially impacted were estimated from clinicaltrials.gov. To identify reporting items considered vital for assessing the impact of COVID-19, we reviewed 35 randomized phase III trials from three top oncology journals published between July and December 2020. For validation, we reviewed 29 phase III trials published between January and December 2021., Results: Our results show that the number of clinical trials being potentially impacted in cancer, cardiovascular diseases, and diabetes is at least 1,484, 535, and 145, respectively. The magnitude of disruption is most significant in oncology trials. Based on the review of 35 trials, a modified checklist with ten new and four modified items covering pandemic's impact on trial conduct, protocol changes, delays, data capture, analysis and interpretation was developed to ensure comprehensive and transparent reporting. Our validation shows that six out of seven applicable reporting items were reported in less than 21% of the articles., Conclusions: Our recommendations were proposed to improve the reporting of randomized clinical trials impacted by COVID-19 and force majeure events that are broadly applicable to different areas of medical research., Competing Interests: Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://atm.amegroups.com/article/view/10.21037/atm-22-2160/coif). THL reports this study partially supported by University Postgraduate Fellowships of HKU Foundation. EEV reports consulting fees and payment/honoraria from AbbVie, AstraZeneca, Beigene, BioNTech, Eli Lilly, EMD Serono, Genentech/Roche, GlaxoSmithKline, Merck and Novatis, outside the submitted work. HP reports personal fees from Genentech, outside the submitted work. The other authors have no conflicts of interest to declare., (2023 Annals of Translational Medicine. All rights reserved.)

Dietary factors show different effects on genetically diverse populations. Scientific research uses gene-environment interaction models to study the effects of dietary factors on genetically diverse populations for lung cancer risk. However, previous study designs have not investigated the degree of type I error inflation and, in some instances, have not corrected for multiple testing. Using a motivating investigation of diet-gene interaction and lung cancer risk, we propose a training and testing strategy and perform real-world simulations to select the appropriate statistical methods to reduce false-positive discoveries. The simulation results show that the unconstrained maximum likelihood (UML) method controls the type I error better than the constrained maximum likelihood (CML). The empirical Bayesian (EB) method can compete with the UML method in achieving statistical power and controlling type I error. We observed a significant interaction between SNP rs7175421 with dietary whole grain in lung cancer prevention, with an effect size (standard error) of -0.312 (0.112) for EB estimate. SNP rs7175421 may interact with dietary whole grains in modulating lung cancer risk. Evaluating statistical methods for gene-diet interaction analysis can help balance the statistical power and type I error.

Background: Weight loss (WL) has been associated with shorter survival in patients with advanced cancer, while obesity has been associated with longer survival. Integrating body mass index (BMI) and WL provides a powerful prognostic tool but has not been well-studied in lung cancer patients, particularly in the setting of clinical trials., Methods: We analysed patient data (n = 10 128) from 63 National Cancer Institute sponsored advanced non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC) trials. Risk matrices were created using BMI and WL percentage, which were divided into 'grades' based on median survival. Relationships between survival, BMI and WL percentage were examined using Kaplan-Meier estimators and Cox proportional hazards (PH) models with restricted cubic splines., Results: For NSCLC, a twofold difference was noted in median survival between the BMI > 28 and WL ≤ 5% group (13.5 months) compared with the BMI < 20 and WL > 5% group (6.6 months). These associations were less pronounced in SCLC. Kaplan-Meier curves showed significant survival differences between grades for both NSCLC and SCLC (log-rank, P < 0.0001). In Stage IV NSCLC, Cox PH analyses with restricted cubic splines demonstrated significant associations between BMI and survival in both WL ≤ 5% (P = 0.0004) and >5% (P = 0.0129) groups, as well as in WL > 5% in Stage III (P = 0.0306). In SCLC, these relationships were more complex., Conclusions: BMI and WL have strong associations with overall survival in patients with advanced lung cancer, with a greater impact seen in NSCLC compared with SCLC. The integration of a BMI/WL grading scale may provide additional prognostic information and should be included in the evaluation of therapeutic interventions in future clinical trials in advanced lung cancer., (© 2022 The Authors. Journal of Cachexia, Sarcopenia and Muscle published by John Wiley & Sons Ltd on behalf of Society on Sarcopenia, Cachexia and Wasting Disorders.)

Background: While the efficacy of immune checkpoint inhibitors (ICIs) is increasingly recognized in advanced gastric cancer (aGC), overall survival (OS) has not been consistently improved across the different randomized controlled trials (RCTs). This meta-analysis aimed to quantify the efficacy and safety of ICI and explore potential predictive tumor tissue biomarkers in aGC., Methods: A random-effect pairwise meta-analysis was used to evaluate the primary outcome of OS. Sensitivity analysis was performed to investigate the effects of ICIs on PD-L1 status, TMB, MSI-H, and the Asian patient population. We extracted the OS Kaplan-Meier curves from the included trials to compare the effect of PD-L1 status on response to ICIs using DigitizeIt 2.5 and Guyot's algorithm., Results: A pairwise meta-analysis of seven RCTs included in this study showed that ICIs were more effective than the comparator in improving OS (pooled HR: 0.84). We demonstrated that PD-1 ICIs were additive when combined with the comparator arm (pooled HR: 0.79). A sensitivity analysis showed that PD-1 ICIs were associated with better OS outcomes in the Asian patient population as monotherapy (pooled HR: 0.66) or in combination with chemotherapy (pooled HR: 0.83). We demonstrated that tumors with PD-L1 ≥1 ( P = 0.02) and PD-L1 ≥10 ( P = 0.006) derived OS benefit from ICI monotherapy. Equally, MSI-H ( P < 0.00001) and TMB-high ( P < 0.0001) tumors derived favorable survival benefits from ICIs., Conclusions and Relevance: The results of this meta-analysis suggest that ICIs result in improved OS outcomes in aGC. The benefits varied with different ethnicities, class of ICI, PD-L1 expression, MSI status, and TMB., Systematic Review Registration: https://www.crd.york.ac.uk/prospero, identifier (CRD42019137829)., Competing Interests: AE disclosure: Advisory role: Pfizer, Bayer, Almac Discovery, Almac Diagnostics Research funding: Almac Diagnostics. MM disclosure: Honoraria: Taiho Pharmaceutical, Eli Lilly/ImClone, Amgen, Roche/Genentech, Merck KGaA, Darmstadt, Germany, MSD Oncology, Bristol Myers Squibb, AstraZeneca/MedImmune, Servier Consulting or Advisory Role: Bayer, Merck Sharp & Dohme, Merck KGaA, Darmstadt, Germany, Amgen, Taiho Pharmaceutical, Nordic Group, Pfizer, Yakult, Roche, Eli Lilly, Servier Research Funding: Amgen Inst, Leap Therapeutics Inst, Merck KGaA, Darmstadt, Germany Inst, Jennerex Inst, AstraZeneca Inst, Merck Sharp & Dohme Inst Travel, Accommodations, Expenses: Amgen, Merck KGaA, Darmstadt, Germany, Roche, Bayer, American Society of Clinical Oncology, German Cancer Society, Merck Sharp & Dohme, European Society for Medical Oncology Kohei Shitara disclosure: Honoraria—Abbvie; Novartis; Yakult Pharmaceutical Consulting or Advisory Role—Abbvie; Amgen; Astellas Pharma; Boehringer Ingelheim; Bristol-Myers Squibb; Daiichi Sankyo; GlaxoSmithKline; Lilly; MSD; Novartis; Ono Pharmaceutical; Pfizer; Taiho Pharmaceutical; Takeda; Novartis; Merck Pharmaceutical Research Funding—Astellas Pharma Inst; Chugai Pharma Inst; Daiichi Sankyo Inst; Dainippon Sumitomo Pharma Inst; Eisai Inst; Lilly Inst; Mediscience Planning Inst; MSD Inst; Ono Pharmaceutical Inst; Taiho Pharmaceutical Inst; Merck Pharmaceutical. NB disclosure: Honoraria—Bristol-Myers Squibb Japan; Ono Pharmaceutical; Taiho Pharmaceutical Research Funding—Ono Pharmaceutical Inst; Takeda Inst. HP disclosure: Personal fees from Genentech outside the submitted work. KL disclosure: Honoraria: Eli Lilly; Bristol-Myers Squibb; Daiichi Sankyo; Taiho Pharmaceutical; Merck Pharmaceutical; Amgen; BAYER; MSD; Sanofi-Aventis Consulting or Advisory Role: Eli Lilly; Bristol-Myers Squibb; Daiichi Sankyo; Taiho Pharmaceutical; Merck Pharmaceutical; Amgen; Bayer; MSD Research Funding: Taiho Pharmaceutical; Bayer; Roche. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Reviewer SM declared a shared affiliation with the author KS to the handling editor at the time of review., (Copyright © 2022 El Helali, Tao, Wong, Chan, Mok, Wu, Shitara, Mohler, Boku, Pang and Lam.)

Background: For patients with locally advanced non-small cell lung cancer (NSCLC), concurrent chemoradiotherapy is the foundational treatment strategy. Adding induction chemotherapy did not achieve a superior efficacy but increased the burden from toxicity. Accordingly, we retrospectively investigated the toxicity patterns through pooling individual patient data of the Cancer and Leukemia Group B (CALGB)/Alliance trials., Methods: We included a total of 637 patients with unresectable stage III NSCLC who received induction chemotherapy with a platinum doublet and concurrent chemoradiotherapy and experienced at least one adverse event (AE) in CALGB 9130, 9431, 9534, 30105, 30106 and 39801 trials. The following toxicity occurrence patterns were evaluated: top 10 most frequent AEs, AE distribution by grade, rate of treatment discontinuation due to AEs, associations of AE occurrence with patient characteristics and treatment phase, the time to the first grade ≥3 AE occurrence and its associations with patient characteristics and treatment phase., Results: The occurrence of AEs was the main reason accounting for treatment discontinuation (60 of 637 among all patients; 18 of 112 patients who experienced the induction phase only; 42 of 525 patients who experienced both phases). All patients experienced a total of 11,786 AEs (grade ≥3: 1,049 of 5,538 in induction phase, 1,382 of 6,248 in concurrent phase). Lymphocytes and white blood count were of top 3 grade ≥3 AEs that patients experienced the most in the either phase. Multivariable analysis found AE occurrence was associated with age ≥65 [any grade: odds ratio (OR) =1.44, 95% confidence interval (CI): 1.12-1.86] and the concurrent phase (grade ≥3: OR =1.86, 95% CI: 1.41-2.47; any grade: OR =1.47, 95% CI: 1.19-1.81). Patients in the concurrent phase were more likely and earlier to develop grade ≥3 AEs than those in the induction phase [hazard ratio (HR) =4.37, 95% CI: 2.52-7.59]., Conclusions: The report provides a better understanding regarding the toxicity occurrence patterns in concurrent chemoradiotherapy after induction chemotherapy., Competing Interests: Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-22-2006/coif). HP has an NIHU01 grant from the FDA, owns stock from Roche, and received personal fees from Genentech, outside the submitted work. XW holds NIH grants but not related to this work. The other authors have no conflicts of interest to declare., (2022 Translational Cancer Research. All rights reserved.)

Background: Proportional hazards (PH) assumption is often violated in cancer immunotherapy studies. Restricted mean survival time (RMST) ratio is a valid metric to quantify the size of treatment effect when non-proportional hazard (NPH) is present. This study investigated the use of RMST ratio and hazard ratio (HR) in studying progression-free survival (PFS) as a surrogate endpoint for overall survival (OS) in non-small cell lung cancer immunotherapy trials., Methods: Trial level data were collected from 14 phase III trials published between 2012 and 2018. A weighted least-square regression (WLSR) was performed to evaluate the trial-level surrogacy. Surrogacy was evaluated via the association between RMST ratios for PFS and OS and between HRs for PFS and OS., Results: Using data extracted from published articles, low to moderate correlation (0.49) between PFS and OS was observed for HR while low correlation (0.35) was observed for RMST ratio. When trials violating PH in PFS were included, more consistent correlations for both HR (0.43) and RMST ratio (0.44) were observed., Conclusions: In summary, the strength of PFS surrogacy for OS depends on whether HR or RMST ratio are chosen. RMST ratio and additional sensitivity analysis should be considered in addition to HR.