Cox proportional hazards regression in small studies of predictive biomarkers.
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
20 06 2024
20 06 2024
Historique:
received:
03
11
2023
accepted:
11
06
2024
medline:
21
6
2024
pubmed:
21
6
2024
entrez:
20
6
2024
Statut:
epublish
Résumé
Predictive biomarkers are essential for personalized medicine since they select the best treatment for a specific patient. However, of all biomarkers that are evaluated, only few are eventually used in clinical practice. Many promising biomarkers may be erroneously abandoned because they are investigated in small studies using standard statistical techniques which can cause small sample bias or lack of power. The standard technique for failure time endpoints is Cox proportional hazards regression with a multiplicative interaction term between binary variables of biomarker and treatment. Properties of this model in small studies have not been evaluated so far, therefore we performed a simulation study to understand its small sample behavior. As a remedy, we applied a Firth correction to the score function of the Cox model and obtained confidence intervals (CI) using a profile likelihood (PL) approach. These methods are generally recommended for small studies of different design. Our results show that a Cox model estimates the biomarker-treatment interaction term and the treatment effect in one of the biomarker subgroups with bias, and overestimates their standard errors. Bias is however reduced and power is increased with Firth correction and PL CIs. Hence, the modified Cox model and PL CI should be used instead of a standard Cox model with Wald based CI in small studies of predictive biomarkers.
Identifiants
pubmed: 38902269
doi: 10.1038/s41598-024-64573-9
pii: 10.1038/s41598-024-64573-9
doi:
Substances chimiques
Biomarkers
0
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
14232Subventions
Organisme : KWF Kankerbestrijding
ID : KWF 10603
Informations de copyright
© 2024. The Author(s).
Références
Lancet, The. 20 years of precision medicine in oncology. Lancet 397(10287), 1791 (2021).
Hoeben, A., Joosten, E. A. J. & van den Beuken-van Everdingen, M. H. J. Personalized medicine: Recent progress in cancer therapy. Cancers 13, 242 (2021).
doi: 10.3390/cancers13020242
pubmed: 33440729
pmcid: 7826530
Ballman, K. V. Biomarker: Predictive or prognostic?. Clin. Oncol. 33(33), 3968–3971 (2015).
Diamandis, E. P. The failure of protein cancer biomarkers to reach the clinic: Why, and what can be done to address the problem?. BMC Med. 10, 87 (2012).
doi: 10.1186/1741-7015-10-87
pubmed: 22876833
pmcid: 3425158
Goossens, N., Nakagawa, S., Sun, X. & Hoshida, Y. Cancer biomarker discovery and validation. Transl. Cancer. Res. 4(3), 256–269 (2015).
pubmed: 26213686
Ou, F.-S., Michiels, S., Shyr, Y., Adjei, A. A. & Oberg, A. L. Biomarker discovery and validation: Statistical considerations. J. Thorac. Oncol. 16(4), 537–545 (2021).
doi: 10.1016/j.jtho.2021.01.1616
pubmed: 33545385
pmcid: 8012218
Altman, D. G., McShane, L. M., Sauerbrei, W. & Taube, S. E. Reporting recommendations for tumor marker prognostic studies (REMARK): Explanation and elaboration. PLoS Med. 9(5), e1001216 (2012).
doi: 10.1371/journal.pmed.1001216
pubmed: 22675273
pmcid: 3362085
Polley, M.-Y.C. et al. Statistical and practical considerations for clinical evaluation of predictive biomarkers. J. Natl. Cancer. Inst. 105(22), 1677–1683 (2013).
doi: 10.1093/jnci/djt282
pubmed: 24136891
pmcid: 3888208
Cox, D. R. Regression models and life-tables. J. R. Stat. Soc. Ser. B. Stat. Methodol. 34(2), 187–220 (1972).
doi: 10.1111/j.2517-6161.1972.tb00899.x
de Boo, L. W. et al. Adjuvant capecitabine-containing chemotherapy benefit and homologous recombination deficiency in early-stage triple-negative breast cancer patients. Br. J. Cancer. 126, 1401–1409 (2022).
doi: 10.1038/s41416-022-01711-y
pubmed: 35124703
pmcid: 9090783
Knauer, M. et al. The predictive value of the 70-gene signature for adjuvant chemotherapy in early breast cancer. Breast Cancer Res. Treat. 120(3), 655–661 (2010).
doi: 10.1007/s10549-010-0814-2
pubmed: 20204499
Kok, M. et al. Estrogen receptor-α phosphorylation at serine-118 and tamoxifen response in breast cancer. J. Natl. Cancer. Inst. 101, 1725–1729 (2009).
doi: 10.1093/jnci/djp412
pubmed: 19940281
Schouten, P. C. et al. Breast cancers with a BRCA1-like DNA copy number profile recur less often than expected after high-dose alkylating chemotherapy. Clin. Cancer. Res. 21(4), 763–770 (2015).
doi: 10.1158/1078-0432.CCR-14-1894
pubmed: 25480832
Vollebergh, M. A. et al. An aCGH classifier derived from BRCA1-mutated breast cancer and benefit of high-dose platinum-based chemotherapy in HER2-negative breast cancer patients. Ann. Oncol. 22(7), 1561–1570 (2011).
doi: 10.1093/annonc/mdq624
pubmed: 21135055
Firth, D. Bias reduction of maximum likelihood estimates. Biometrika 80(1), 27–38 (1993).
doi: 10.1093/biomet/80.1.27
Heinze, G. & Schemper, M. A solution to the problem of monotone likelihood in Cox regression. Biometrics 57, 114–119 (2001).
doi: 10.1111/j.0006-341X.2001.00114.x
pubmed: 11252585
Heinze, G. The application of Firth’s procedure to Cox and logistic regression, Technical Report 10/1999, update in January 2001, Section of Clinical Biometrics, Department of Medical Computer Sciences University of Vienna (2001).
Morris, T. P., White, I. R. & Crowther, M. J. Using simulation studies to evaluate statistical methods. Stat. Med. 38(11), 2074–2102 (2019).
doi: 10.1002/sim.8086
pubmed: 30652356
pmcid: 6492164
Bender, R., Augustin, T. & Blettner, M. Generating survival times to simulate Cox proportional hazards models. Stat. Med. 24, 1713–1723 (2005).
doi: 10.1002/sim.2059
pubmed: 15724232
Fleiss, J. L., Levin, B. & Paik, M. C. Statistical Methods for Rates and Proportions 3rd ed. (Wiley, 2003).
doi: 10.1002/0471445428
Venzon, D. J. & Moolgavkar, S. H. A method for computing profile-likelihood-based confidence intervals. J. R. Stat. Soc. C Appl. Stat. 37(1), 87–94 (1988).
Heinze, G., Ploner, M., Jiricka, L. & Steiner, G. Cox regression with Firth’s penalized likelihood. Package coxphf version 1.13.4 (2023). https://cran.r-project.org/web/packages/coxphf/index.html .
Vittinghoff, E. & Bauer, D. C. Case-only analysis of treatment-covariate interactions in clinical trials. Biometrics 62, 769–776 (2006).
doi: 10.1111/j.1541-0420.2006.00511.x
pubmed: 16984319
Liu, Y., Jiang, W. & Chen, B. E. Testing for treatment-biomarker interaction based on local partial-likelihood. Stat. Med. 34, 3516–3530 (2015).
doi: 10.1002/sim.6563
pubmed: 26084729
Dai, J. Y., Zhang, X. C., Wang, C.-Y. & Kooperberg, C. Augmented case-only designs for randomized clinical trials with failure time endpoints. Biometrics 72, 30–38 (2016).
doi: 10.1111/biom.12392
pubmed: 26347982
Jiang, S., Chen, B. & Tu, D. Inference on treatment-covariate interaction based on a nonparametric measure of treatment effects and censored survival data. Stat. Med. 35, 2715–2725 (2016).
doi: 10.1002/sim.6907
pubmed: 26887976
Sollfrank, L., Linn, S. C., Hauptmann, M. & Jóźwiak, K. A scoping review of statistical methods in studies of biomarker-related treatment heterogeneity for breast cancer. BMC Med. Res. Methodol. 23, 154 (2023).
doi: 10.1186/s12874-023-01982-w
pubmed: 37386356
pmcid: 10308726
SAS Institute Inc. SAS/STAT® 15.1 User’s Guide. Cary, NC: SAS Institute Inc. (2018).
Heinze, G. FC06: A SAS® macro for Cox regression with Firth’s penalization. Technical report 1/2006 (2006).
Ioannidis, J. P. A. Why most published research findings are false. PLoS Med. 2(8), e124 (2005).
doi: 10.1371/journal.pmed.0020124
pubmed: 16060722
pmcid: 1182327
Garcia-Closas, M. & Lubin, J. H. Power and sample size calculations in case-control studies of gene-environmental interactions: Comments on different approaches. Am. J. Epidemiol. 149, 689–693 (1999).
doi: 10.1093/oxfordjournals.aje.a009876
pubmed: 10206617