RE: Progression of Frailty in Survivors of Childhood Cancer: A St. Jude Lifetime Cohort Report

We read the article by Delaney et al. (1) with interest. This study was prominent because it clarified some important factors regarding the prediction of frailty and a longitudinal association between frailty and mortality in childhood cancer survivors using a large cohort from a tertiary hospital in the United States. The authors highlighted the modifiable risk factors that might guide effective preventive interventions for frailty and death, such as physical activity and sedentary time. However, we assume that the relationship between these risk factors and frailty remains associational and not causal. Given our viewpoint, we would like to discuss what this study should have done for an approach to causation.

First, this study (1) should have presented cause-specific mortality, similar to a previously published study from the same database (2). It remains unknown how frailty contributes to the subsequent death of childhood cancer survivors. Current evidence shows that the 2 leading causes of death in childhood cancer survivors are the recurrence or progression of primary cancers and secondary malignancies (3). Suppose the association between frailty and cause-specific mortality is clarified. In that case, we may speculate the role of frailty in the causal pathway (eg, frailty might be a mediator between primary or secondary malignancies and death, or frailty directly causes disability and death among childhood cancer survivors).

Second, the study by Delaney et al. (1) should have included factors that could explain the mechanisms between candidate factors and frailty in the multivariable model for frailty. The results would have been more informative if the authors could clarify the factors that can account for specific etiologies of frailty, such as the alkylating agent dose. One example is nonalcoholic fatty liver disease (NAFLD). NAFLD frequently occurs among cancer survivors (4) and is associated with sarcopenia, which is closely related to frailty (5). Thus, NAFLD may be a candidate to explain the underlying mechanism of frailty.

Third, the study (1) should have constructed another multivariable model to examine the causal relationship between lifestyle factors and mortality. In this study, the purpose of multivariable analysis for frailty was to explore the risk factors that predict frailty. Consequently, physical activity and sedentary time were statistically significant predictors of frailty. This study did not establish a causal relationship between these factors and mortality. Therefore, the construction of another multivariable model is essential. In the model, all confounding factors based on the causal diagram should be appropriately adjusted for (6). We suggest socioeconomic status as a candidate confounding factor—for example, household equivalent income or academic attainment.

Taken together, caution should be exercised when speculating on the causal relationship between lifestyle factors and mortality because of the abovementioned three points. Given the drastic improvement in the prognosis of patients with childhood cancer (7), there is no doubt that focusing on frailty and associated lifestyle is essential for all health-care professionals working in childhood cancer care. We believe that the study by Delaney et al. (1) sheds light on frailty, its associated factors, and the association between frailty and mortality, all of which are important for pediatric oncology. Further research that compensates for the issues mentioned in this article is required.

Funding

None.

Notes

Role of the funder: Not applicable.

Disclosures: The authors have no disclosures.

Author contributions: Junya Makiyama: Conceptualization, investigation, visualization, writing—original draft, writing—review and editing. Ryo Momosaki: Conceptualization, investigation, validation, writing—review and editing. Toshifumi Yodoshi: Conceptualization, investigation, validation, writing—review and editing. Takeshi Fujieda: Conceptualization, investigation, validation, writing—review and editing. Akihiro Ozaka: Conceptualization, investigation, validation, writing—review and editing. Atsushi Takayama: Conceptualization, investigation, visualization, validation, writing—original draft, writing—review and editing. Takashi Yoshioka: Conceptualization, investigation, methodology, project administration, supervision, validation, visualization, writing—original draft, writing—review and editing.

Data Availability

No new data were generated or used for this correspondence.