Pelvic fractures and changes in bone mineral density after radiotherapy for cervical, endometrial, and vaginal cancer: A prospective study of 239 women

Abstract

BACKGROUND: Advances in radiotherapy (RT) have led to improved oncologic outcomes for women with gynecologic cancers; however, the long-term effects and survivorship implications need further evaluation. The purpose of this study was to determine the incidence of pelvic fractures and changes in bone mineral density (BMD) after pelvic RT. METHODS: Two hundred thirty-nine women who had pelvic RT for cervical, endometrial, or vaginal cancer between 2008 and 2015 were prospectively studied. BMD scans and biomarkers of bone turnover were obtained at the baseline and 3 months, 1 year, and 2 years after RT. Imaging studies were assessed for pelvic fractures for up to 5 years. Patients with osteopenia, osteoporosis, or pelvic fractures at any point were referred to the endocrinology service for evaluation and treatment. RESULTS: The median age at diagnosis was 51 years; 132 patients (56%) were menopausal. The primary diagnoses were cervical (63.6%), endometrial (30.5%), and vaginal cancer (5.9%). Sixteen patients (7.8%; 95% confidence interval, 4.5%-12.4%) had pelvic fractures with actuarial rates of 3.6%, 12.7%, and 15.7% at 1, 2, and 3 years, respectively. Fractures were associated with baseline osteoporosis (P < .001), higher baseline bone-specific alkaline phosphatase (P < .001), and older age (P = .007). The proportion of patients with osteopenia/osteoporosis increased from 50% at the baseline to 58%, 59%, and 70% at 3 months, 1 year, and 2 years, respectively. CONCLUSIONS: A high proportion of women had significant decreases in BMD after pelvic RT, with 7.8% diagnosed with a pelvic fracture. BMD screening and pharmacologic intervention should be strongly considered for these high-risk women.

Assessing risk factors with information beyond P value thresholds: Statistical significance does not equal clinical importance

“…Observational studies published in Cancer may seek to identify risk factors for an outcome of interest by using multivariable models or multiple univariable models if the sample size is limited. 1,2 Among multiple candidate predictor variables, emphasis is commonly placed on those explanatory variables achieving statistical significance, which is generally defined as a P value below a threshold of .05 (P < .05); within this group, those with the smallest P values are often highlighted. This article illustrates ways in which this practice can be misleading and endorses confidence intervals (CIs), graphical approaches, and prespecified effects to facilitate a more complete understanding of findings.…”

“…To illustrate some of these ideas in practice, consider a recent example from Cancer describing the incidence of pelvic fracture and associated risk factors after radiotherapy for cervical, endometrial, and vaginal cancer. 1 The example was chosen solely to illustrate the widespread practice of focusing on those candidate risk factors with the smallest P values. Salcedo et al 1 describe a total of 16 events recorded in up to 200 women.…”

“…Power calculations should always be made before the analyses of these types and reported in the article according to journal guidelines. 16 The stated goal of Salcedo et al 1 of "determining … associated risk factors" implies that those risk factors with an association have a reasonable chance of being identified. Salcedo et al highlight those predictors with P < .05 and focus on those with the smallest P values.…”

“…The example was chosen solely to illustrate the widespread practice of focusing on those candidate risk factors with the smallest P values. Salcedo et al 1 describe a total of 16 events recorded in up to 200 women. Their Table 4 reports results for candidate risk factors based on 15 univariable Cox models.…”

“…Forest plot of HRs and 95% CIs for the binary risk factors reported in Salcedo et al 1 (2020). The dotted line at 1 indicates no association (HR, 1), and the dashed line at 1.5 shows a 50% increase in the hazard (HR, 1.5).…”