Abstract
Introduction: Survivors of childhood cancer are at risk for neurocognitive impairment. The pattern and degree of such impairment in childhood non-Hodgkin lymphoma (NHL) survivors has not been previously assessed.
Methods: Adult survivors of childhood NHL participating in the SJLIFE Cohort Study, ≥10 years from diagnosis, and ≥18 years old were recruited for neurocognitive testing, with age-adjusted z-scores compared to community controls and normative data. Cumulative treatment exposures were abstracted from medical records. Multivariable modeling, adjusting for age, gender, and time from diagnosis, was used to calculate relative risks (RR) and 95% confidence intervals (CI).
Results: 187 NHL survivors (59% of eligible) completed neurocognitive testing. Survivors were a median [range] age at diagnosis of 10 [2 - 21] years, median time from diagnosis 26 [11 - 48] years, and median age at evaluation 35 [19 - 58] years. Fifty-six (30%) received cranial radiation (CRT), 70 (37%) high-dose methotrexate (HD-MTX), 40 (21%) high-dose cytarabine (HD-ARAC), and 151 (81%) intrathecal (IT) chemotherapy. Survivors had full scale intelligence (mean = -0.1, SD = 1.0, range: -2.7 - 2.0), memory (mean= -0.2, SD = 1.0, range: -4.2 - 1.6), and attention (mean = 0, SD = 0.7, range: -2.5 - 1.4) within normal limits. Survivors demonstrated worse executive function, processing speed, and academics compared to community controls and normative data (Table 1). In multivariable models, there were no statistically significant associations with worse academics, processing speed, executive function and behavior rating and CRT, HD-MTX, HD-ARAC, or IT chemotherapy. Survivors reported more depression and somatization compared to community controls (p's < 0.001) and normative data (p's < 0.01). CRT (p = 0.003) was associated with improved somatization while longer time from diagnosis (p = 0.01) was associated with worse somatization. Survivors reported worse emotional quality of life compared to community controls (p = 0.005) and normative data (p < 0.001). Following adjustment for neurocognitive impairment, self-reported neurobehavioral problems (shift and emotional control) were associated with worse emotional quality of life (vitality, social functioning, mental health) (p's < 0.001). There was no statistically significant difference between survivors and controls for educational attainment and full time employment. Survivors with worse academic (RR = 1.5, 95% CI: 1.2 - 1.7) and executive function (RR = 1.1, 95% CI: 1.0 - 1.2) were more likely to not graduate from college. Those with lower processing speed were more likely to be unemployed or work only part time (RR = 1.3, 95% CI: 1.1 - 1.4).
Conclusions: Adult survivors of childhood NHL experience impaired neurocognitive function, mental health, and quality of life, which at 26 years post-diagnosis is not related to original treatment exposures. Associations between impairment and chronic health conditions should be explored.
No relevant conflicts of interest to declare.
