Lifestyle and subsequent meningioma in childhood cancer survivors: A report from the St. Jude Lifetime Cohort study
Abstract
Background
Lifestyle is associated with meningioma risk in the general population.
Aims
We assessed longitudinal associations between lifestyle-associated factors and subsequent meningiomas in childhood cancer survivors.
Methods and results
Childhood cancer survivors age ≥18 years in the St. Jude Lifetime Cohort Study were evaluated for body composition, self-reported physical activity, cardiopulmonary fitness, muscle strength, smoking, and alcohol consumption at baseline. Time to first meningioma analyses were performed, adjusted for sex, age at diagnosis and baseline assessment, treatment decade, and childhood cancer treatment exposures. The study included 4,072 survivors (47% female; [mean (SD)] 9 (6) years at diagnosis; 30 (8.5) years at the start of follow-up, with 7.0 (3.3) years of follow-up). 30% of the participants were survivors of acute lymphoblastic leukemia and 29% of the participants had received cranial radiation. During follow-up, 90 participants developed ≥1 meningioma, of whom 73% were survivors of acute lymphoblastic leukemia, with cranial radiation being the strongest risk factor (relative risk [RR] 29.7, 95% confidence interval [CI] 10.6-83.2). Muscle strength assessed by knee extension was associated with a lower risk of developing a meningioma in the adjusted analyses (RR 0.5, 95% CI 0.2-1.0, p = 0.04 for quartiles 3-4 vs. 1). No other lifestyle-associated variable was associated with subsequent meningioma.
Conclusion
Independent of cranial radiation, muscle strength was associated with a lower risk of developing a subsequent meningioma in childhood cancer survivors.
1 BACKGROUND
Low physical activity and increased body fat are associated with a higher risk of developing meningioma in the general population,1, 2 in addition to affecting women more than men.3 Childhood cancer survivors are at increased risk of developing meningioma, attributed to cranial radiotherapy.4 These meningiomas generally occur decades after the childhood cancer diagnosis and may cause both neurological sequelae and death.4 Whether lifestyle can modify the increased risk of meningioma in childhood cancer survivors is not known. This study aimed to assess longitudinal associations between lifestyle and subsequent meningiomas in childhood cancer survivors included in the St Jude Lifetime Cohort study (SJLIFE).5
2 METHODS
This study was an observational analysis within SJLIFE. The assessments in SJLIFE have been described previously.5, 6 Participants enrolled in the St. Jude Lifetime Cohort who were ≥18 years of age and ≥5 years from childhood cancer diagnosis at the time of assessment, and had undergone physical performance and/or body composition assessment were eligible for this study.
2.1 Exposures
2.1.1 Cardiopulmonary fitness
Cardiopulmonary fitness was evaluated by meters walked from the 6-min walk test.7-9 The results were categorized into quartiles for the analyses and the lowest quartile was used as reference in the statistical analyses.
2.1.2 Muscular strength
Muscular strength was evaluated by knee extension strength (Nm/kg) and handgrip strength testing (kg). Isokinetic knee extension strength was measured as peak torque from five repetitions at 60° per second. The maximum value was used for analyzing both knee extension strength and handgrip strength. The results were categorized into quartiles, with the third and fourth quartiles combined. The lowest quartile was used as reference in the analyses.
2.1.3 Body composition
Body composition was evaluated with body mass index (BMI; kg/m2) from the clinical evaluation and body fat percentage from DEXA evaluations. We categorized BMI as underweight (<18.5), normal (≥18.5 and <25), overweight (≥25 and <30), and obesity (≥30). Body fat percentage was categorized according to the Obesity Medicine Association 2022 guidelines.10 For men, 25%–29% was categorized as pre-obesity and ≥30% as obesity. For women, 30%–34% was categorized as pre-obesity and ≥35% as obesity. Normal weight was used as reference for both BMI- and DEXA-defined variables.
2.1.4 Physical activity
Physical activity was self-reported as moderate-to-vigorous physical activity minutes per day, which was then converted into metabolic equivalent task (MET) hours per week and categorized into 0–3, 3–6, and ≥6 METh/week. 0–3 MET hours per week was used as reference.
2.1.5 Smoking
Self-reported smoking, categorized as never, previous, or current smoking. Current smoking was used as reference in the analyses.
2.1.6 Alcohol consumption
Risky drinking was defined as >3 drinks per day or >7 per week for women, >4 per day or >14 per week for men. Risky drinking was used as reference in the analyses.
2.2 Outcome
Time to first meningioma during survivorship in adult age was the outcome, including both benign and malignant meningiomas. Only meningiomas occurring after the baseline assessment were included. All SJLIFE participants have ascertained follow-up of subsequent malignancies through medical record abstraction and yearly follow-up with the St Jude Cancer Registry.11 There is no routine CNS-imaging included in the research evaluations in SJLIFE. Hence, all meningiomas were diagnosed from clinically motivated follow-up. Meningioma diagnoses were ascertained from histopathological diagnosis where available and otherwise from imaging.
2.3 Treatment exposures
Medical record abstraction for eligible SJLIFE participants includes abstraction of all chemotherapy received, including cumulative doses.11 In this study, we included dichotomized treatment exposures (yes/no) to anthracyclines, alkylating agents, epipodophyllotoxins, platinum, and brain radiation.
2.4 Statistical methods
Piecewise exponential models estimated risk for meningioma as relative risk (RR) with 95% confidence interval (CI) adjusted for sex, age and calendar year at childhood cancer diagnosis, age at baseline assessment, chemotherapy exposures (alkylating agents, platinum, epipodophyllotoxins, anthracyclines), and cranial radiation. P-values from the piecewise exponential models for each comparison are provided. Follow-up started 1 year after the baseline campus visit and stopped at the time of first meningioma, date of last contact, or death, whichever came first. Each model included one of the exposures of interest (cardiopulmonary fitness, body composition, physical activity, strength, smoking, or risky drinking) and the predefined confounders known to affect the risk for meningiomas.
3 RESULTS
The study included 4072 survivors, 1925 (47%) female, 1133 (29%) previous/current smokers, 1435 (37%) risky drinkers, 1384 (34%) with obesity; mean (standard deviation [SD]) age was 8.7 (5.7) years at diagnosis, and 30 (8.4) years at baseline SJLIFE evaluation (Table 1). Subsequent length of follow-up was 7.1 (3.3) years after assessment. At least one meningioma was experienced by 90 (2.2%) survivors, of whom 66 (73%) had been treated for acute lymphoblastic leukemia (ALL), and 85 (96%) had received cranial radiation therapy (Table 1).
| Characteristic | Developed meningioma | No meningioma | Total |
|---|---|---|---|
| Number of participants | 90 | 3982 | 4072 |
| Sex, female | 50 (56%) | 1875 (47%) | 1925 (47%) |
| Age at baseline, years, mean (SD) | 36.2 (7.7) | 30.0 (8.4) | 30.1 (8.5) |
| Years of follow-up, mean (SD) | 4.5 (2.7) | 7.1 (3.3) | 7.0 (3.3) |
| Race/ethnicity | |||
| Non-Hispanic Caucasian | 82 (91%) | 3175 (80%) | 3257 (80%) |
| Non-Hispanic African American | 8 (9%) | 615 (15%) | 623 (15%) |
| Hispanic | 0 (0%) | 114 (3%) | 114 (3%) |
| Other | 0 (0%) | 78 (2%) | 78 (2%) |
| Educational attainment | |||
| Up to high school or GED | 29 (33%) | 1164 (31%) | 1193 (31%) |
| Training after high school or some college | 31 (35%) | 1345 (36%) | 1376 (36%) |
| College graduate or postgraduate | 28 (32%) | 1206 (32%) | 1234 (32%) |
| Missing | 2 | 267 | 269 |
| Household income, per year | |||
| <$20 000 | 35 (50%) | 1357 (43%) | 1392 (43%) |
| $20 000–<$59 999 | 19 (27%) | 1034 (33%) | 1053 (33%) |
| ≥ <$60 000 | 16 (23%) | 747 (24%) | 763 (24%) |
| Missing | 20 | 844 | 864 |
| Body mass index, kg/m2, mean (SD) | 32.4 (9.8) | 28.1 (7.1) | 28.2 (7.2) |
| Underweight | 2 (2%) | 148 (4%) | 150 (4%) |
| Normal weight | 17 (19%) | 1388 (35%) | 1405 (35%) |
| Overweight | 26 (29%) | 1106 (28%) | 1132 (28%) |
| Obesity | 45 (50%) | 1339 (34%) | 1384 (34%) |
| Body fat percentage, %, mean (SD) | 36 (8) | 32 (9) | 32 (9) |
| Non obese | 5 (12%) | 899 (28%) | 904 (28%) |
| Pre obesity | 5 (12%) | 744 (23%) | 749 (23%) |
| Obesity | 31 (76%) | 1576 (49%) | 1607 (49%) |
| Missing | 49 | 763 | 812 |
| Cardiorespiratory fitness | |||
| 6-min walk test, meters walked, mean (SD) | 523 (106) | 556 (104) | 556 (104) |
| Missing | 4 | 159 | 163 |
| Grip strength, kg, mean (SD) | 34 (12) | 39 (13) | 39 (13) |
| Missing | 0 | 8 | 8 |
| Knee extension, Nm/kg, mean (SD) | 120 (57) | 142 (56) | 141 (56) |
| Missing | 11 | 449 | 460 |
| Self-reported physical activity, METh/week, median (IQR) | 6.0 (0.0–19.1) | 9.0 (0.0–24.0) | 9.0 (0.0–24.0) |
| Missing | 4 | 235 | 239 |
| Smoking history | |||
| Never | 69 (79%) | 2715 (71%) | 2784 (71%) |
| Past | 11 (13%) | 453 (12%) | 464 (12%) |
| Current | 7 (8%) | 662 (17%) | 669 (17%) |
| Missing | 3 | 152 | 155 |
| Heavy/risky drinking of alcohol | 18 (21%) | 1417 (38%) | 1435 (37%) |
| Missing | 4 | 220 | 224 |
| Healthy lifestyle score | |||
| Poor | 41 (46%) | 1905 (48%) | 1946 (48%) |
| Moderate | 36 (40%) | 1311 (33%) | 1347 (33%) |
| Healthy | 13 (14%) | 766 (19%) | 779 (19%) |
| Decade of primary cancer diagnosis | |||
| 1960–1970 | 8 (9%) | 111 (3%) | 119 (3%) |
| 1971–1980 | 42 (47%) | 583 (15%) | 625 (15%) |
| 1981–1990 | 28 (31%) | 1064 (27%) | 1092 (27%) |
| 1991–2000 | 11 (12%) | 1346 (34%) | 1357 (33%) |
| 2001 | 1 (1%) | 878 (22%) | 879 (22%) |
| Primary childhood cancer diagnosis | |||
| ALL | 66 (73%) | 1140 (29%) | 1206 (30%) |
| Central nervous system tumor | 12 (13%) | 541 (14%) | 553 (14%) |
| Other | 12 (13%) | 2301 (57%) | 2313 (57%) |
| Age at primary cancer diagnosis, years | 7 (5) | 9 (6) | 9 (6) |
| Cancer treatment | |||
| Received cranial radiation | 85 (96%) | 1088 (28%) | 1173 (29%) |
| Received alkylating agents | 54 (60%) | 2305 (58%) | 2359 (58%) |
| Received anthracyclines | 41 (46%) | 2303 (58%) | 2344 (58%) |
| Received platinum | 10 (11%) | 559 (14%) | 569 (14%) |
| Received epipodophyllotoxins | 41 (46%) | 1422 (36%) | 1463 (36%) |
| Hematopoietic stem cell transplant | 6 (7%) | 319 (8%) | 325 (8%) |
- Note: Numbers are n (%) if not stated otherwise.
- Abbreviation: MET, metabolic equivalent task.
3.1 Univariate analyses
In the univariate analyses, self-reported physical activity was not associated with the subsequent risk of developing meningioma (RR 0.7, 95% CI 0.5–1.2 for high vs. low categories), while both BMI- (RR 2.5, 95% CI 1.5–4.4 for obesity vs. normal weight) and DEXA-assessed (RR 6.3, 95% CI 2.5–15.6 for obesity vs. normal weight) obesity were associated with an increased risk. Cardiorespiratory fitness (RR 0.4, 95% CI 0.2–0.7 for high vs. low quartiles), grip strength (RR 0.5, 95% CI 0.3–0.7 for quartiles 3–4 vs. 1), knee extension strength (RR 0.3, 95% CI 0.2–0.5 for quartiles 3–4 vs. 1), smoking (RR 1.8, 95% CI 1.1–3.1 for never smoking vs. current smoking) and risky drinking (RR 2.3, 95% CI 1.4–3.9 for no risky drinking vs. risky drinking) were associated with lower risk of developing a meningioma (Table 2).
| Comparison | Univariate analyses | Multivariable analyses | ||
|---|---|---|---|---|
| RR (95% CI) | P for comparison | RR (95% CI) | P for comparison | |
| Body mass index | ||||
| Overweight (25–29.9) vs. normal | 1.8 (1.0–3.3) | 0.06 | 1.3 (0.7–2.4) | 0.42 |
| Obesity (≥30) vs. normal | 2.5 (1.5–4.4) | 0.001 | 1.2 (0.7–2.1) | 0.59 |
| Body fat percentage | ||||
| Pre-obesity vs. normal | 2.4 (0.8–6.9) | 0.11 | 1.1 (0.4–3.3) | 0.81 |
| Obesity vs. normal | 6.3 (2.5–15.6) | <0.001 | 1.8 (0.7–4.7) | 0.23 |
| Self-reported physical activity | ||||
| 3–6 vs. 0–3 METh/week | 0.7 (0.4–1.3) | 0.26 | NA | NA |
| ≥6 vs. 0–3 METh/week | 0.7 (0.5–1.2) | 0.20 | NA | NA |
| 6-min walk test | ||||
| Quartile 2 vs. 1 | 0.6 (0.4–1.1) | 0.09 | 0.8 (0.5–1.5) | 0.53 |
| Quartile 3 vs. 1 | 0.5 (0.3–0.9) | 0.02 | 0.8 (0.4–1.4) | 0.43 |
| Quartile 4 vs. 1 | 0.4 (0.2–0.7) | 0.002 | 0.7 (0.4–1.4) | 0.33 |
| Grip strength | ||||
| Moderate vs. low | 0.4 (0.2–0.7) | 0.002 | 0.6 (0.3–1.1) | 0.12 |
| High vs. low | 0.5 (0.3–0.7) | 0.001 | 0.8 (0.3–1.7) | 0.49 |
| Knee extension | ||||
| Moderate vs. low | 0.5 (0.3–0.9) | 0.03 | 0.7 (0.4–1.1) | 0.14 |
| High vs. low | 0.3 (0.2–0.5) | <0.001 | 0.5 (0.2–1.0) | 0.04 |
| Never smoker | 1.8 (1.1–3.1) | 0.02 | 1.6 (0.9–2.9) | 0.11 |
| Absence of risky drinking | 2.3 (1.4–3.9) | 0.002 | 1.3 (0.7–2.3) | 0.36 |
- Note: One lifestyle variable was included in each model. All models were adjusted for sex, age at diagnosis and baseline assessment, treatment decade, and childhood cancer treatment exposures.
- Abbreviations: BMI, body mass index; CNS, central nervous system; MET, metabolic equivalent task; RR, relative risk.
3.2 Multivariable models
As expected, cranial radiation exposure was the strongest risk factor for developing meningiomas in this childhood cancer survivor population (RR 29.7, 95% CI 10.6–83.2). In our multivariable models including treatment variables, knee extension strength was the only lifestyle variable associated with the risk of developing a subsequent meningioma (RR 0.5, 95% CI 0.2–1.0 for quartiles 3–4 vs. 1, Table 2). None of the meningiomas occurred in participants with growth hormone replacement at baseline or neurofibromatosis types 1 or 2. For BMI- (RR 1.2, 95% CI 0.7–2.1) and fat percentage-defined (RR 1.8, 95% CI 0.7–4.7) obesity, the adjusted estimates changed considerably from the univariate analyses, indicating confounding by treatment exposures. For cardiorespiratory fitness (RR 0.7, 95% CI 0.4–1.4) and grip strength (RR 0.8, 95% CI 0.3–1.7), the difference in estimates between univariate and adjusted analyses was less dramatic, albeit there were no associations.
4 DISCUSSION
In this observational study of 4072 childhood cancer survivors, we report a protective association between knee extension strength and the risk of developing meningioma, independent of cranial radiation therapy. Previous studies have reported increased risk of meningioma with increasing body fat and physical inactivity in the general population.1, 2
Our study could not confirm associations between self-reported physical activity or BMI and meningioma. There are several possible explanations for this difference. Childhood cancer survivors are at increased risk of sarcopenic obesity and BMI is a poor measure for identifying obesity in childhood cancer survivors, compared to DEXA.12 This might reflect the difference in RR between BMI- and DEXA-assessed obesity, albeit none of them were significant after adjustment for treatment exposures and other potential confounders. Self-reported physical activity measures have relatively low sensitivity in the general population.13 Objectively assessed cardiorespiratory and muscular fitness are more sensitive measures and are improved mainly by exercise training and less by low-intensity PA,14 and have been reported to be stronger predictors of other health outcomes, such as all-cause mortality.14 A recently published study showed associations between cardiorespiratory fitness and site-specific cancers across several organ systems but did not look at meningiomas specifically.15 In our univariate analyses, both favorable body composition and all measures of cardiorespiratory and musculoskeletal fitness were associated with decreased risk of developing meningioma. Our multivariable analyses could only confirm associations for knee extension strength. This might be explained by residual confounding from cranial radiation treatment but may also reflect true associations between, for example, exercise and meningiomas. This should be further assessed in studies from other cohorts and in future studies with longer follow-up. Since cranial radiation increases the risk of both meningioma and obesity, it can be hard to discriminate possible direct effects of obesity on meningioma. It is also possible that some of the results, for example, for DEXA-assessed body composition suffered from low statistical power. Most survivors who developed a meningioma did not undergo DEXA scans and a longer follow-up might result in significant associations. The fact that all associations for body composition and fitness were in the expected direction, albeit not significant, could imply that a longer follow-up with more meningiomas during follow-up would yield further significant associations. There is a long-time latency between childhood cancer treatment and meningioma onset. Our population was followed from a mean of 22–28 years after diagnosis, corresponding to the peak in meningioma incidence in other studies.4
This study was performed within SJLIFE, with clinically ascertained health outcomes and objectively assessed fitness and body composition. To our knowledge, our study is the first observational cohort study assessing the association between lifestyle factors and meningioma in childhood cancer survivors. The main limitation is the relatively short follow-up of participants, resulting in a limited number of events, limiting statistical power. The absence of histopathological diagnosis for the majority of meningiomas is a limitation. However, performing neurosurgical operations to obtain tissue samples when not clinically motivated cannot be done. The absence of systematically performed CNS imaging in all participants in SJLIFE is a limitation. However, this would have either exposed participants to unnecessary radiation from CT scans or required them to undergo MRI scans and would need to be justified. These results from a single observational study need to be confirmed in other populations, preferably with diverse ethnicities or from different nations, to strengthen any assumptions on causality. However, they are supported by similar associations in the general population.
In conclusion, independent of treatment exposures, knee extension strength was associated with a lower risk of developing meningioma in childhood cancer survivors. It remains to be shown whether this reflects a causal effect.
AUTHOR CONTRIBUTIONS
Aron Onerup: Conceptualization (lead); funding acquisition (supporting); investigation (equal); methodology (equal); writing – original draft (lead); writing – review and editing (lead). Sedigheh Mirzaei S.: Conceptualization (equal); formal analysis (equal); investigation (equal); methodology (equal); supervision (equal); validation (equal); writing – review and editing (equal). Shalini Bhatia: Formal analysis (lead); investigation (equal); methodology (supporting); validation (equal); writing – review and editing (equal). Megan E. Ware: Investigation (supporting); methodology (supporting); writing – review and editing (supporting). Lenat Joffe: Investigation (supporting); methodology (supporting); writing – review and editing (supporting). Lucie M. Turcotte: Investigation (supporting); methodology (supporting); writing – review and editing (supporting). Chelsea G. Goodenough: Investigation (supporting); methodology (supporting); writing – review and editing (supporting). Yadav Sapkota: Investigation (supporting); methodology (supporting); writing – review and editing (supporting). Stephanie B. Dixon: Conceptualization (supporting); investigation (equal); methodology (supporting); writing – review and editing (equal). Matthew D. Wogksch: Investigation (supporting); methodology (supporting); writing – review and editing (supporting). Matthew J. Ehrhardt: Conceptualization (supporting); investigation (supporting); methodology (supporting); writing – review and editing (supporting). Gregory T. Armstrong: Conceptualization (supporting); investigation (supporting); methodology (supporting); project administration (equal); resources (equal); writing – review and editing (supporting). Melissa M. Hudson: Conceptualization (supporting); funding acquisition (equal); investigation (supporting); methodology (supporting); resources (supporting); writing – review and editing (supporting). Kirsten K. Ness: Conceptualization (equal); funding acquisition (equal); investigation (equal); methodology (equal); project administration (equal); resources (equal); supervision (lead); writing – review and editing (equal).
ACKNOWLEDGMENTS
None.
CONFLICT OF INTEREST STATEMENT
The authors have stated explicitly that there are no conflicts of interest in connection with this article.
ETHICS STATEMENT
The St. Jude Lifetime Cohort study was approved by the Institutional Review Board at St. Jude Children's Research Hospital, with current approval date December 3, 2023.
Open Research
DATA AVAILABILITY STATEMENT
The data supporting this study's findings are openly available at: https://www.stjude.cloud/research-domains/cancer-survivorship. Data specific to this paper will be uploaded to https://zenodo.org concomitant with the publication of the manuscript.
