TL;DR
This study identifies distinct profiles of behavioral, social, and emotional symptoms in adolescent survivors of childhood cancer, revealing that these symptoms often co-occur and are linked to specific cancer treatments and long-term physical side effects. Understanding these symptom profiles is crucial for tailoring mental health interventions to improve outcomes for this vulnerable population.
What Problem Does This Paper Address?
Adolescence is a critical period for emotional and social development, and individuals who have survived childhood cancer face unique challenges. While it's known that psychological symptoms can co-occur in the general adolescent population, there's been a lack of detailed understanding regarding the specific profiles of these comorbid symptoms in childhood cancer survivors. This gap makes it difficult to effectively screen for and treat these issues, potentially leading to poorer long-term mental health and well-being for survivors. The paper aims to fill this gap by identifying these symptom profiles and exploring their associations with treatment exposures and physical late effects.
Methods
The researchers utilized data from the Childhood Cancer Survivor Study (CCSS), a large, multi-institutional project. Their primary analytical technique was Latent Profile Analysis (LPA). LPA is a statistical method used to identify unobserved subgroups (or "profiles") within a population based on patterns in their measured variables. In this case, the measured variables were scores on the Behavior Problems Index (BPI), which assesses various symptom domains like anxiety/depression, headstrong behavior, attention deficit, social withdrawal, and antisocial behavior.
After identifying these symptom profiles, the study employed multivariable multinomial logistic regression to examine the associations between these profiles and various factors, including:
- Cancer treatment exposures: such as cranial radiation therapy (CRT), chemotherapy agents (methotrexate, cytarabine, anthracyclines), and corticosteroids.
- Physical late effects: like obesity, growth hormone deficiency, scarring, sensory impairment, cancer-related pain, and migraines.
- Demographic covariates: age, sex, race/ethnicity, and income.
Analyses were stratified by cranial radiation therapy (CRT) exposure to account for its significant impact on neurocognitive and behavioral outcomes. To address multiple comparisons, a false discovery rate approach was used.
Datasets
The study used data from the Childhood Cancer Survivor Study (CCSS). This dataset includes over 3,893 adolescent survivors of childhood cancer (aged 12-17 years at baseline) who were treated between 1970 and 1999. Parent-reported behavioral data from the Behavior Problems Index (BPI) was the primary outcome measure.
Key Insights and Ideas
- Symptom profiles are distinct and common: The study successfully identified distinct patterns of symptom co-occurrence, moving beyond simply looking at individual symptoms. This highlights that these issues rarely occur in isolation.
- CRT significantly shapes symptom profiles: Survivors treated with cranial radiation therapy (CRT) showed different symptom profiles compared to those who did not receive CRT. Notably, an "externalizing symptoms only" profile was absent in the CRT group, suggesting CRT's impact on brain development influences the manifestation of these behaviors.
- Physical late effects are strong predictors: Beyond direct treatment effects, long-term physical consequences like obesity, sensory impairments, and cancer-related pain were significantly associated with increased odds of specific symptom profiles. This underscores the interconnectedness of physical and mental health in survivors.
- Comorbidity is the norm, not the exception: The findings reinforce that in adolescent psychopathology, comorbidity is common, and survivors are not immune to this trend. This has direct implications for how we approach screening and intervention.
Key Results
- Four symptom profiles were identified in the no-CRT group:
- Well-adjusted (no significant symptoms): ~70%
- Externalizing symptoms (headstrong behavior, attention problems): ~16%
- Internalizing symptoms (anxiety/depression, social withdrawal, attention problems): ~9%
- Global symptoms (externalizing and internalizing): ~5%
- Three symptom profiles were identified in the CRT group:
- Well-adjusted: ~63%
- Internalizing symptoms: ~31%
- Global symptoms: ~6% (An externalizing-only profile was not observed in the CRT group.)
- Treatment associations:
- Receiving ≥ 30 Gy of CRT increased the risk of internalizing (OR 1.7) and global symptoms (OR 3.2).
- In the no-CRT group, nondexamethasone corticosteroids were associated with a higher risk of externalizing symptoms (OR 1.9).
- High-dose IV methotrexate (≥ 4.3 g/m²) was associated with a lower risk of internalizing symptoms (OR 0.53) and externalizing symptoms (OR 0.5) in the no-CRT group.
- Physical late effects associations (No CRT group):
- Overweight/obesity was linked to increased internalizing (OR 2.0) and global symptoms (OR 1.8).
- Sensory impairment was associated with a 2.5-fold increased risk of internalizing symptoms.
- Cancer-related pain and migraines were associated with increased internalizing symptoms.
Limitations and Future Work
The authors acknowledge several limitations:
- Parent-reported data: Behavioral outcomes were based on parent reports, which can sometimes differ from adolescent self-reports, particularly for internalizing symptoms.
- Cross-sectional design: The study design does not allow for definitive conclusions about causality or temporal relationships between exposures and symptom development.
- Normative sample: Using a sibling comparison group as a normative sample might be a limitation, as siblings may also be affected by the family's experience with cancer.
- Generalizability: While the CCSS is large, results might not be fully generalizable to survivors treated with the most recent therapeutic approaches.
Future research directions suggested include:
- Examining developmental differences: Further investigation into how symptom profiles evolve across early versus later adolescence.
- Longitudinal studies: Tracking symptom profiles over time to understand their stability and trajectory.
- Functional outcomes: Investigating the impact of these symptom profiles on long-term functional outcomes and quality of life.
- Incorporating survivor self-report: Future studies should ideally include both parent and survivor perspectives.
- Targeted screening: The findings strongly support the need for comprehensive screening that goes beyond just internalizing symptoms, particularly for survivors with specific treatment exposures (like CNS-directed therapies) or physical late effects.
Summary updated