Dear Editor,
We read with great interest the article by Dr. Schotten et al. [1]. In this retrospective single-center cohort study, the safety (treatment-related toxicity) and efficacy (overall survival) of yttrium-90 radioembolization between nonalcoholic fatty liver disease (NAFLD)-associated and hepatitis B virus (HBV)-associated hepatocellular carcinoma (HCC) were compared. The results of this study showed that, at a comparable toxicity profile in the 87 patients with NAFLD-associated and the 62 patients with HBV-associated HCC, median overall survival after yttrium-90 radioembolization for patients with NAFLD-associated HCC was similar to patients with HBV-associated HCC (11.1 vs. 9.3 months, p = 0.38). Herein, we would like to raise the following comments.
In this study, the baseline characteristics of patients with NAFLD-associated HCC and HBV-associated HCC were identified and compared. As shown in Table 1 of this study, there were significant or marginally significant differences in many clinical variables between these two groups, including mean age (p < 0.01), European Cooperative Oncology Group criteria (p = 0.07), the presence of cirrhosis (p = 0.01), Child-Pugh score (p = 0.07), size of major nodule (p = 0.05), etc., which suggests an unbalanced enrollment due to these confounding variables between the two groups. As we all know, propensity score methods have been increasingly used to reduce the selection bias in estimating treatment effects and allow investigators to reduce the likelihood of confounding when analyzing nonrandomized, observational data [2, 3]. Actually, one similar comparative study between NAFLD-associated and HBV-associated HCC in patients undergoing liver resection (but not yttrium-90 radioembolization) has previously been reported, and propensity score matching analysis was well performed in that study [4]. Although the conclusions of both studies indicated that regardless of yttrium-90 radioembolization or liver resection, the safety and efficacy between NAFLD-associated and HBV-associated HCC were comparable, the latter study [4], from the perspective of statistical reliability, is much better than the previous one [1]. Herein, we propose propensity score analysis, adopted in Schotten et al.'s study [1]. In addition, multivariate Cox regression analysis of overall survival should also be performed in the entire cohort to identify whether there was an independent survival difference between patients with NAFLD-associated and HBV-associated HCC.
Disclosure Statement
The authors have no conflicts of interest to declare.
Financial Support
None.
Author Contributions
Conception: Lei Liang, Li-Yang Sun, Yong-Kang Diao, Tian Yang. Manuscript preparation: Lei Liang, Li-Yang Sun. Critical revision: Yong-Kang Diao, Tian Yang. All the authors reviewed the paper and approved the final version. Lei Liang and Li-Yang Sun contributed equally to this work.
References
- 1.Schotten C, Bechmann LP, Manka P, Theysohn J, Dechêne A, El Fouly A, et al. NAFLD-Associated Comorbidities in Advanced Stage HCC Do Not Alter the Safety and Efficacy of Yttrium-90 Radioembolization. Liver Cancer. 2019 Nov;8((6)):491–504. doi: 10.1159/000501484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Thomas L, Li F, Pencina M. Using Propensity Score Methods to Create Target Populations in Observational Clinical Research. JAMA. 2020 Jan;323((5)):466. doi: 10.1001/jama.2019.21558. [DOI] [PubMed] [Google Scholar]
- 3.Haukoos JS, Lewis RJ. The Propensity Score. JAMA. 2015 Oct;314((15)):1637–8. doi: 10.1001/jama.2015.13480. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Yang T, Hu LY, Li ZL, Liu K, Wu H, Xing H, et al. Liver Resection for Hepatocellular Carcinoma in Non-alcoholic Fatty Liver Disease: a Multicenter Propensity Matching Analysis with HBV-HCC. J Gastrointest Surg. 2019 Jan;24((2)):320–9. doi: 10.1007/s11605-018-04071-2. [DOI] [PubMed] [Google Scholar]