Non-parametric individual treatment effect estimation for survival data with random forests

2019 ◽  
Author(s):  
Sami Tabib ◽  
Denis Larocque
Keyword(s):  
Random Forest ◽  
Survival Time ◽  
Treatment Effect ◽  
Survival Data ◽  
Supplementary Information ◽  
Individual Treatment ◽  
Cancer Data ◽  
Mean Survival Time ◽  
The Mean ◽  
The Individual

Abstract Motivation Personalized medicine often relies on accurate estimation of a treatment effect for specific subjects. This estimation can be based on the subject’s baseline covariates but additional complications arise for a time-to-event response subject to censoring. In this paper, the treatment effect is measured as the difference between the mean survival time of a treated subject and the mean survival time of a control subject. We propose a new random forest method for estimating the individual treatment effect with survival data. The random forest is formed by individual trees built with a splitting rule specifically designed to partition the data according to the individual treatment effect. For a new subject, the forest provides a set of similar subjects from the training dataset that can be used to compute an estimation of the individual treatment effect with any adequate method. Results The merits of the proposed method are investigated with a simulation study where it is compared to numerous competitors, including recent state-of-the-art methods. The results indicate that the proposed method has a very good and stable performance to estimate the individual treatment effects. Two examples of application with a colon cancer data and breast cancer data show that the proposed method can detect a treatment effect in a sub-population even when the overall effect is small or nonexistent. Availability and implementation The authors are working on an R package implementing the proposed method and it will be available soon. In the meantime, the code can be obtained from the first author at [email protected]. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Estimation of Heterogeneous Restricted Mean Survival Time Using Random Forest

2021 ◽  
Vol 11 ◽  
Author(s):  
Mingyang Liu ◽  
Hongzhe Li
Keyword(s):  
Random Forest ◽  
Survival Time ◽  
Survival Data ◽  
Survival Function ◽  
Parametric Models ◽  
The Cancer Genome Atlas ◽  
Data Set ◽  
Cancer Data ◽  
Mean Survival Time ◽  
Restricted Mean Survival Time

Estimation and prediction of heterogeneous restricted mean survival time (hRMST) is of great clinical importance, which can provide an easily interpretable and clinically meaningful summary of the survival function in the presence of censoring and individual covariates. The existing methods for the modeling of hRMST rely on proportional hazards or other parametric assumptions on the survival distribution. In this paper, we propose a random forest based estimation of hRMST for right-censored survival data with covariates and prove a central limit theorem for the resulting estimator. In addition, we present a computationally efficient construction for the confidence interval of hRMST. Our simulations show that the resulting confidence intervals have the correct coverage probability of the hRMST, and the random forest based estimate of hRMST has smaller prediction errors than the parametric models when the models are mis-specified. We apply the method to the ovarian cancer data set from The Cancer Genome Atlas (TCGA) project to predict hRMST and show an improved prediction performance over the existing methods. A software implementation, srf using R and C++, is available at https://github.com/lmy1019/SRF.

scholarly journals Survival Analysis of Treatment Efficacy in Comparative Coronavirus Disease 2019 Studies

2020 ◽  
Author(s):  
Zachary R McCaw ◽  
Lu Tian ◽  
Dae Hyun Kim ◽  
A Russell Localio ◽  
Lee-Jen Wei
Keyword(s):  
Survival Analysis ◽  
Mortality Rate ◽  
Survival Time ◽  
Treatment Effect ◽  
Treatment Efficacy ◽  
Hazard Ratio ◽  
Time Difference ◽  
Short Term ◽  
Mean Survival Time ◽  
The Mean

Abstract For survival analysis in comparative coronavirus disease 2019 trials, the routinely used hazard ratio may not provide a meaningful summary of the treatment effect. The mean survival time difference/ratio is an intuitive, assumption-free alternative. However, for short-term studies, landmark mortality rate differences/ratios are more clinically relevant and should be formally analyzed and reported.

scholarly journals Extrapolating Parametric Survival Models in Health Technology Assessment Using Model Averaging: A Simulation Study

2021 ◽  
Vol 41 (4) ◽  
pp. 476-484
Author(s):  
Daniel Gallacher ◽  
Peter Kimani ◽  
Nigel Stallard
Keyword(s):  
Survival Time ◽  
Survival Data ◽  
Mean Squared Error ◽  
Model Averaging ◽  
Information Criterion ◽  
Health Technology ◽  
Weighted Averaging ◽  
Health Technologies ◽  
Mean Survival Time ◽  
Restricted Mean Survival Time

Previous work examined the suitability of relying on routine methods of model selection when extrapolating survival data in a health technology appraisal setting. Here we explore solutions to improve reliability of restricted mean survival time (RMST) estimates from trial data by assessing model plausibility and implementing model averaging. We compare our previous methods of selecting a model for extrapolation using the Akaike information criterion (AIC) and Bayesian information criterion (BIC). Our methods of model averaging include using equal weighting across models falling within established threshold ranges for AIC and BIC and using BIC-based weighted averages. We apply our plausibility assessment and implement model averaging to the output of our previous simulations, where 10,000 runs of 12 trial-based scenarios were examined. We demonstrate that removing implausible models from consideration reduces the mean squared error associated with the restricted mean survival time (RMST) estimate from each selection method and increases the percentage of RMST estimates that were within 10% of the RMST from the parameters of the sampling distribution. The methods of averaging were superior to selecting a single optimal extrapolation, aside from some of the exponential scenarios where BIC already selected the exponential model. The averaging methods with wide criterion-based thresholds outperformed BIC-weighted averaging in the majority of scenarios. We conclude that model averaging approaches should feature more widely in the appraisal of health technologies where extrapolation is influential and considerable uncertainty is present. Where data demonstrate complicated underlying hazard rates, funders should account for the additional uncertainty associated with these extrapolations in their decision making. Extended follow-up from trials should be encouraged and used to review prices of therapies to ensure a fair price is paid.

scholarly journals Laryngeal Neuroendocrine Carcinomas: A Retrospective Study of 14 Cases

2015 ◽  
Vol 2015 ◽  
pp. 1-4 ◽  
Author(s):  
Yingying Zhu ◽  
Liming Gao ◽  
Yunxiao Meng ◽  
Wenwen Diao ◽  
Xiaoli Zhu ◽  
...  
Keyword(s):  
Survival Time ◽  
Neuroendocrine Carcinoma ◽  
Clinical Manifestations ◽  
Small Cell ◽  
Atypical Carcinoid ◽  
Small Cell Neuroendocrine Carcinoma ◽  
Typical Carcinoid ◽  
Mean Survival Time ◽  
The Mean ◽  
Neuroendocrine Carcinomas

Laryngeal neuroendocrine carcinomas (LNECs) are rare and highly heterogeneous which present a wide spectrum of pathological and clinical manifestations. Fourteen patients with histologically demonstrated LNEC were collected and analyzed retrospectively. The 14 cases were classified into 3 subtypes: typical carcinoid in 2, atypical carcinoid in 5, and small cell neuroendocrine carcinoma in 7. The mean survival time of the 14 patients in this study was 112.5 months (95% CI, 81.5–143.6). Surgeries were performed for 2 patients of typical carcinoid, and they were alive with no evidence of recurrence after 24 and 47 months of follow-ups. Patients in the atypical carcinoid group were treated with surgeries and postoperative radiotherapy. After 58.4 months of follow-ups (range: 9–144), 2 patients showed no evidence of disease and 1 was lost to follow-up after 72 months. The other 2 patients died of other unrelated diseases. In the small cell neuroendocrine carcinoma group, a combination of chemotherapy and radiotherapy was applied. The mean survival time was 79.7 months (95% CI, 37.9–121.4), and the 5-year survival rate was 53.6%. In conclusion, the clinical behaviors, treatment protocols, and prognosis are different for each subtype of LNECs.

Effect of Intravenous Iron in Experimental Traumatic Shock

1956 ◽  
Vol 186 (3) ◽  
pp. 554-556
Author(s):  
Donn L. Smith ◽  
Irvin I. Kibbey ◽  
Max E. Bierwagen ◽  
J. R. Cruse
Keyword(s):  
Heavy Metal ◽  
Survival Time ◽  
Metal Toxicity ◽  
Intravenous Iron ◽  
Traumatic Shock ◽  
Survival Times ◽  
Albino Rat ◽  
Liver Iron ◽  
Mean Survival Time ◽  
The Mean

Intravenous administration of colloidal saccharated iron oxide prior to intestinal traumatization in the albino rat resulted in a significant reduction of the mean survival time. Sodium gold thiosulfate and colloidal manganese hydroxide employed in the same manner did not significantly alter mean survival times. ACTH and cortisone did not modify the deleterious effects of iron in experimental traumatic shock. A decrease in soluble liver iron was observed when traumatization followed the injection of iron. It was concluded that the reduction of mean survival time in iron injected, traumatized animals was due to a specific action of iron and is not the result of generalized heavy metal toxicity.

scholarly journals Big End Double-Layer Stents for the Treatment of Gastric Outlet Obstruction Caused by Stomach Cancer

2019 ◽  
Vol 2019 ◽  
pp. 1-6
Author(s):  
Ding Shi ◽  
Dong Wu ◽  
Yongpan Liu ◽  
Feng Ji ◽  
Yinsu Bao
Keyword(s):  
Gastric Cancer ◽  
Survival Time ◽  
Stomach Cancer ◽  
Double Layer ◽  
Gastric Outlet Obstruction ◽  
Outlet Obstruction ◽  
Metal Stents ◽  
Mean Survival Time ◽  
Distal Gastric Cancer ◽  
The Mean

Objectives. This study is aimed at evaluating the efficacy and safety of the big end double-layer uncovered self-expanding metal stents (SEMS) for the treatment of gastric outlet obstruction (GOO) caused by distal stomach cancer.Methods. Seventy three patients receiving big end double-layer uncovered SEMS for the treatment of GOO caused by distal gastric cancer will be included in this multicenter prospective clinical trial. The main outcome measures included the functional outcome, the complications, the reinterventional rates, the average treatment charges, and the mean survival time. Monthly telephone calls were needed to assess the food intake until the patients died.Results. The technical and the clinical success rates were 98.6%. The stent obstruction caused by tumor ingrowth was observed in one patient (1.4%). The incidence of food impaction was 2.9% (2/70) and the reinterventional rate was 4.3% (3/70). However, stent migration and obstruction caused by overgrowth were not observed. No perforation and severe bleeding were observed. The median cost of endoscopic stenting and total hospitalization (including reinterventions) for the big end double-layer uncovered SEMS in this study was $2945 and $3408, respectively. The mean survival time was 212.5 days.Conclusions. The placement of big end double-layer uncovered SEMS is a safe and effective modality and has the potential to be one of the options for the treatment of GOO caused by the distal gastric cancer.

scholarly journals A Combination of UTMD-Mediated HIF-1α shRNA Transfection and TAE in the Treatment of Hepatic Cancer

2019 ◽  
Vol 2019 ◽  
pp. 1-8 ◽  
Author(s):  
Yangying Liao ◽  
Haibo Luo ◽  
Zhizhong He ◽  
Yongpei Kuang ◽  
Peifen Chen ◽  
...  
Keyword(s):  
Survival Time ◽  
Western Blotting ◽  
Tumor Size ◽  
Statistical Difference ◽  
Antitumor Effect ◽  
Control Group ◽  
Cancer Tissue ◽  
Hepatic Cancer ◽  
Mean Survival Time ◽  
The Mean

To explore the antitumor effect of hypoxia-inducible factor-1α short hairpin RNA (HIF-1α shRNA) delivered by ultrasound targeted microbubble destruction (UTMD) and transcatheter arterial embolization (TAE) on rats with hepatic cancer. After the models of transplantation hepatoma were established, Wistar rats were randomly divided into 4 groups: Control group, UTMD group, TAE group, and UTMD+TAE group. Contrast-enhanced ultrasound (CEUS) was used to monitor tumor size on day 14 after four different treatments. Western blotting and immunohistochemistry were applied to measure the protein level of HIF-1α and VEGF in the hepatic cancer tissue. In comparison with UTMD+TAE group (21.25±10.68 days), the mean survival time was noticeably shorter in the Control group and TAE group (13.02±4.30 days and 15.03±7.32 days) (p<0.05, respectively). There was no statistical difference between UTMD+TAE group and UTMD group of the mean survival time (p>0.05). In addition, our results proved that the tumor sizes in UTMD+TAE group were obviously smaller than those in other groups (p<0.05, respectively). By CEUS, we clearly found that the tumor size was the smallest on day 14 in the UTMD+TAE group. The western blotting and immunohistochemistry results proved that the protein levels of HIF-1α and VEGF in UTMD+TAE group were obviously lower than those in TAE group and Control group on days 7 and 14 (p<0.05, respectively). However, there was no statistical difference between UTMD+TAE group and UTMD group (p>0.05). In this study we tried to explore the antitumor effect through a combination of UTMD-mediated HIF-1α shRNA transfection and TAE on rats with hepatic cancer. Our results showed that UTMD-mediated HIF-1α shRNA transfection and TAE can obviously silence HIF-1α and VEGF expression, thereby successfully inhibiting the growth of the tumor.

Different effects of hypothermia on two syndromes of positive acceleration

1963 ◽  
Vol 18 (2) ◽  
pp. 387-392 ◽  
Author(s):  
E. R. Stiehm
Keyword(s):  
Survival Time ◽  
Cardiac Failure ◽  
Cerebral Circulation ◽  
The Other ◽  
Mean Survival Time ◽  
Cardiac Syndrome ◽  
Positive Acceleration ◽  
The Mean

The tolerance to high positive (headward) acceleration at levels of 20–80 G was studied in normal and hypothermic rats. Normal rats have a mean survival time of 680 sec at 20 G, but this decreases to 105 sec at 30 G. Only slight decreases are noted at higher G, suggesting that 30 G is the point at which cerebral circulation is interrupted. Hypothermia at 22.5 C decreases acceleration tolerance at 20 G, but markedly increases acceleration tolerance at 30 G and above. At 40 G this effect is maximal, and the mean survival time for hypothermic rats is 252 sec, compared to 75 sec for controls, an increase of 236%. These different effects of hypothermia on acceleration tolerance suggest that there are two distinct syndromes of physiologic failure during positive acceleration. One is a “cardiac” syndrome, occurring at 20 G and below, characterized by partial maintenance of cerebral circulation until cardiac failure ensues, and the other is a “cerebral” syndrome, occurring at 30 G and above, characterized by immediate interruption of cerebral circulation and respiratory paralysis. Submitted on May 17, 1962

Relationship Between Survival Rate and Survival-Time of Rabbits, Oryctolagus-Cuniculus (L), Challenged With Myxoma Virus

1995 ◽  
Vol 43 (3) ◽  
pp. 303 ◽  
Author(s):  
I Parer
Keyword(s):  
Survival Time ◽  
Small Groups ◽  
Regression Line ◽  
Survival Rates ◽  
Myxoma Virus ◽  
Survival Times ◽  
Mean Survival Time ◽  
The Past ◽  
Survival Percentage ◽  
The Mean

The mean survival times of small groups of rabbits challenged with myxoma virus have been used to estimate survival rates and to allocate virulence grades to field strains of myxoma virus. The slope of the regression Line relating survival percentage to mean survival time in days was shown to be less steep than has been previously estimated. This overestimation of the regression slope has, in the past, resulted in most field strains of myxoma virus being allocated to the Grade III level of virulence when allocation to Grade I would have been more appropriate.

Sign in / Sign up

Export Citation Format

Share Document