Survival Analysis for Pancreatic Cancer Patients using Cox-Proportional Hazard (CPH) Model

Aditya Chakraborty; Chris P. Tsokos

doi:10.34257/GJMRFVOL21IS3PG29

Authors

Aditya Chakraborty
Chris P. Tsokos

Keywords:

pancreatic cancer, cox-PH model, pancreatic survival function

Abstract

Pancreatic cancer is comparatively rare but extremely lethal. In the United States, pancreatic cancer is the 4 th leading cause of cancer death, and in Europe, it is the 6 th . Though Pancreatic cancer remains incurable if detected late, research into improving the therapeutic strategy has increased significantly in recent years. However, it is ambiguous if sustained improvements have been achieved by identifying the most prominent risk factors responsible for cancer. In this article, we studied the survival times of 677 pancreatic cancer patients with fifteen risk factors. The semi-parametric Cox proportional hazard (CPH) model was used to examine the covariate effect taking into account all of the statistically significant risk factors and their significant twoway interactions. A careful and rigorous assessment of the risk factors based on the AIC of the stepwise selection technique revealed seven risk factors, and ten interaction terms are statistically significantly contributing to the survival times. The final Cox-PH model was wellvalidated and satisfied all the key assumptions. The identified risk factors and their interactions are ranked according to the prognostic effect on the survival time based on the hazard ratio. We found the most contributing risk factor is the combined effect of patients with emphysema and cancer stage regional with a hazard ratio (HR) = 8.84.

Downloads

How to Cite

Survival Analysis for Pancreatic Cancer Patients using Cox-Proportional Hazard (CPH) Model. (2021). Global Journal of Medical Research, 21(F3), 29-46. https://doi.org/10.34257/GJMRFVOL21IS3PG29

Download Citation

References

(null) Ric. 4. Patient K., diagnosis: congenital bilateral clubfoot, severe degree. a - before treatment (age 10 days); b - after conservative treatment (age 9 months).. (1).

Stage (Categorical) (X 2 ): Pancreatic Cancer Stages, categorized as a) localized, b) regional, and c) distant.

(null) Risk of hemorrhagic stroke from aspirin use: Does risk outweigh the benefit?.

Ibuprofen (Categorical) (X 4 ): Does the person use Ibuprofen Regularly?.

(2020) Number of First Degree Relatives (Affected).

): Did the patient ever have diabetes?.

(2005) Existing coronary heart disease is undiagnosed in half of women who have a first heart attack.

(2003) Financial Aspects of Emphysema and Emphysema Surgery. 440-451.

(null) Table 1: Percentages of the analysed categorical variables divided according to patient sex.. (9).

Bmi numeric) (X 10 ): Current Body Mass Index (BMI) at Baseline.

The total number of years the patient smoked.

Eshwar Karupakula (2023) Sigmoid diverticulitis with rectal diverticulosis.

Gareth Iacobucci (2024) Vaping: One million adults in England who never regularly smoked now use e-cigarettes. q2160.

Donghui Li, Keping Xie, Robert Wolff, James Abbruzzese (2004) Pancreatic cancer. 363(9414), 1049-1057.

Audrey Vincent, Joseph Herman, Rich Schulick, Ralph Hruban, Michael Goggins (2011) Pancreatic cancer. 378(9791), 607-620.

J Kleeff, M Korc, M Apte (2016) Pancreatic cancer. 2, 16022.

J Ferlay (2012) cancer incidence and mortality worldwide: IARC CancerBase. 11.

David Cox (1972) Regression Models and Life-Tables. 34(2), 187-202.

Muditha Perera, Chris Tsokos (2018) A Statistical Model with Non-Linear Effects and Non-Proportional Hazards for Breast Cancer Survival Analysis. 07(01), 65-89.

Junichi Asano, Akihiro Hirakawa, Chikuma Hamada (2014) Assessing the prediction accuracy of cure in the Cox proportional hazards cure model: an application to breast cancer data. 13(6), 357-363.

X Yong, C Tsokos (2009) PROBABILISTIC COMPARISON OF SURVIVAL ANALYSIS MODELS USING SIMULATION AND CANCER DATA.

Xudong Du, Mier Li, Ping Zhu, Ju Wang, Lisha Hou, Jijie Li, Hongdao Meng, Muke Zhou, Cairong Zhu (2018) Comparison of the flexible parametric survival model and Cox model in estimating Markov transition probabilities using real-world data. 13(8), e0200807.

David Kleinbaum, Mitchel Klein (2012) The Cox Proportional Hazards Model and Its Characteristics. 97-159.

J Kalbeisch, R Prentice (1980) The statistical analysis of failure time data. By J.D. Kalbfleisch and R.L. Prentice. John Wiley & Sons, Inc., New York, 1980. xi + 321 pp. U.S. $31.50, C $40.35. ISBN 0‐471‐05519‐0. 10(1), 64-66.

Brody Tom (2011) Clinical Trials: Study Design, Endpoints and Biomarkers, Drug Safety, and FDA and ICH Guidelines. 165-168.

H Akaike (1974) A new look at the statistical model identification. 19(6), 716-723.

A Sashegyi, D Ferry (2017) On the Interpretation of the Hazard Ratio and Communication of Survival Benefit. 22(4), 484-486.

Case Douglas, Gretchen Kimmick, Electra Paskett, Kurt Lohman, Robert Tucker (2002) Interpreting Measures of Treatment Effect in Cancer Clinical Trials. 7, 181-187.

Wing Wong (1986) Theory of Partial Likelihood. 14(1), 88-123.

Luis Fernandez (1986) Non-parametric maximum likelihood estimation of censored regression models. 32(1), 35-57.

Patricia Grambsch, Terry Therneau (1994) Proportional hazards tests and diagnostics based on weighted residuals. 81(3), 515-526.

A Winnett (2001) Miscellanea. A note on scaled Schoenfeld residuals for the proportional hazards model. 88(2), 565-571.

L Mamudu, C Tsokos, O Oluwaseun (2020) Survival Analysis of Multiple Myeloma Cancer (MMC) Using the Cox-Proportional Hazard Model. 5(7), 147.

T Therneau, P Grambsch, T Fleming (1990) Martingale-based residuals for survival models. 77(1), 147-160.

Dominique Michaud (2004) Epidemiology of Pancreatic Cancer. 59, 471-487.

Aditya Chakraborty, Chris Tsokos (2021) Parametric and Non-Parametric Survival Analysis of Patients with Acute Myeloid Leukemia (AML). 11(01), 126-148.

Aditya Chakraborty, Dr. P. Tsokos (2021) A Real Data-Driven Analytical Model to Predict Happiness. 8(3), 45-61.