Concentration, working speed and memory: cognitive problems in young childhood cancer survivors and their siblings

Abstract

Background: Cognitive problems can have a negative effect on a person's education, but little is known about cognitive problems in young childhood cancer survivors (survivors). This study compared cognitive problems between survivors and their siblings, determined if cognitive problems decreased during recent treatment periods and identified characteristics associated with the presence of a cognitive problem in survivors.

Methods: As part of the Swiss Childhood Cancer Survivor Study, a questionnaire was sent to all survivors, aged 8-20 years, registered in the Swiss Childhood Cancer Registry, diagnosed at age <16 years, who had survived ≥ 5 years. Parent-reported (aged 8-15 years) and self-reported (aged 16-20 years) cognitive problems (concentration, working speed, memory) were compared between survivors and siblings. Multivariable logistic regression was used to identify characteristics associated with cognitive problems in survivors.

Results: Data from 840 survivors and 247 siblings were analyzed. More often than their siblings, survivors reported problems with concentration (12% vs. 6%; P = 0.020), slow working speed (20% vs. 8%; P = 0.001) or memory (33% vs. 15%; P < 0.001). Survivors from all treatment periods were more likely to report a cognitive problem than were siblings. Survivors of CNS tumors (OR = 2.82 compared to leukemia survivors, P < 0.001) and those who had received cranial irradiation (OR = 2.10, P = 0.010) were most severely affected.

Conclusion: Childhood cancer survivors, even those treated recently (2001-2005), remain at risk to develop cognitive problems, suggesting a need to improve therapies. Survivors with cognitive problems should be given the opportunity to enter special education programs.

Keywords: childhood cancer; cognitive outcomes; cohort study; questionnaire survey; siblings; survivors.

Fig. 1

Study participants.

Fig. 2

Reporting of any cognitive problem in survivors compared to siblings, overall and stratified by period of diagnoses, treatment and diagnoses. Odds ratios are retrieved from four multivariable logistic regressions: 1. SURVIVORS: OR for being a survivor compared to being a sibling; 2. PERIOD OF TREATMENT: OR for having had the diagnoses in the period 2001–2005, 1996–2000 or 1989–1995 compared to being a sibling; 3. TREATMENT: OR for having had surgery only, chemotherapy (may include surgery), radiotherapy (may include surgery and/or chemotherapy or bone marrow transplantation (may include surgery, and/or chemotherapy and/or radiotherapy) compared to siblings; 4. DIAGNOSES: OR for having a diagnosis of leukemia, lymphoma, CNS tumor, neuroblastoma, retinoblastoma, renal/hepatic tumor, bone tumor, soft tissue sarcoma, germ cell tumor, other tumor or Langerhans cell hystiocytosis compared to being a sibling. All regressions were controlled for gender, age at survey, migration background, language region, parents’ education. The siblings’ population is standardized on age, gender, migration background, language region and parents’ education according to the survivor population. Family clusters were used to adjust standard errors for the fact that siblings were not independent from survivors.