Beyond Risk-Based Stratification: Impacts of Processing Speed and Executive Function on Adaptive Skills in Adolescent and Young Adult Cancer Survivors

Abstract

Purpose:

The number of adolescent and young adult (AYA) survivors of childhood cancer is increasing, and the impacts of therapy on their daily lives are not well understood. Adaptive functions are required for age-appropriate interactions and day to day functioning, but are reduced in AYA survivors. Work in other pediatric populations suggests that additional neurocognitive skills may influence adaptive function and, thus, quality of life and personal attainment of AYA cancer survivors.

Methods:

Retrospective medical records review examined neurocognitive data from 139 AYA survivors. Hierarchical linear regression examined age at diagnosis, use of central nervous system (CNS) radiation, verbal intelligence, processing speed, and executive function as predictors of adaptive functioning domains.

Results:

AYA survivors exhibited weaknesses in all domains of adaptive functioning compared to normative reference values (Cohen's d = 0.660–0.864), as well as in processing speed (Cohen's d = 0.791) and metacognitive executive functioning (Cohen's d = 0.817). Processing speed and executive function provided substantial improvements in prediction of adaptive functioning beyond that of age at diagnosis and use of CNS-directed radiation therapy. Taken together these variables explained 37.1% of variability in adaptive conceptual skills, 26.1% in adaptive social skills, and 27.1% of adaptive practical skills.

Conclusions:

Intelligence, processing speed, and executive function significantly contribute to adaptive function scores in AYA cancer survivors and impact domains that are important to self-sufficiency and quality of life. Attention to neurocognitive function in all AYA cancer survivors is recommended in addition to referral for neuropsychological evaluation and tailoring interventions to address executive and adaptive functioning.

Introduction

The annual incidence of cancers in children, adolescents, and young adults below age 39 years has increased to 75.7 per 100,000 persons.¹ Fortunately, survivorship rates for these patients have been increasing, and the 5-year survival rate for children, adolescents, and young adults (AYAs) diagnosed with cancer has increased to 83%–85%,¹ which means that the number of AYA survivors of cancer is increasing and will continue to do so. The improved survivorship is attributed to multiple factors, including an increase in the intensity of therapies. Unfortunately, the therapies required to treat cancer can create lifelong neurocognitive effects that significantly impact quality of life, cognitive functioning, and overall personal achievement.^2–4

Survivors of childhood cancers exhibit significant differences in academic and neurocognitive functioning, with deficits in multiple cognitive domains, including numerical processing, executive function, and adaptive function, compared to those without a history of cancer.^5–8 Children who receive central nervous system (CNS) directed chemotherapy or radiation therapy later exhibit dose-dependent cognitive toxicities from therapy.⁵ Because of the unique developmental expectations facing AYAs, treatment-related cognitive impairments create significant challenges for development of age-appropriate skills and self-sufficiency that are hallmarks of this developmental stage. Neurocognitive impacts of therapy have been reported as the second most concerning self-identified symptom after treatment by AYAs,³ and these cognitive effects create tremendous challenges to the transition back to work or school post-therapy.⁹ As a result, AYAs have survivorship and psychosocial supportive needs that are unique from other age groups.¹⁰

Adaptive function skills include a person's ability to care for themselves, navigate effectively within home and community settings, and interact with others at an age-appropriate level; these skills reflect how well a person functions independently and may be negatively impacted by chemotherapy and radiation therapy.^5,8,11 Changes in adaptive function have been shown to reflect a dose-dependent effect of craniospinal radiation therapy in cancer survivors.^8,12 AYA survivors treated with chemotherapy show diminished cognitive and educational attainment compared to those without a history of cancer.¹³ Young adult survivors of childhood cancer and those with worse neurocognitive impairment are also less likely to live independently compared to healthy siblings.¹⁴

Executive functions are proposed contributors to adaptive skills given the need to initiate timely actions and regulate responses in real-world settings and are common areas of weakness in survivors of childhood cancers.^15–18 Executive functions are processes that are required to control one's own behavior and emotions and include attention regulation, inhibitory control, working memory, and cognitive control and flexibility. Executive function has been identified as a predictor of adaptive function in children with a variety of noncancer populations,^20–22 but less work has examined this association in AYA survivors of childhood cancer. As survival rates increase, there is a need to better understand what life will look like after cancer treatment in AYAs as their executive and adaptive demands increase. Furthermore, few studies exist that clarify how known areas of neurocognitive risk (e.g., reasoning ability, processing speed, and executive function) relate to functional adaptive outcomes in AYA survivors.

The purpose of the present study is to describe the adaptive function of AYA survivors and examine prediction of adaptive function from clinical demographic variables (age at diagnosis, type of diagnosis, and use of CNS-directed radiation therapy) and cognitive function (verbal intelligence and executive function).

Methods

A retrospective cross-sectional study was conducted at the Kennedy Krieger Institute and Johns Hopkins Children's Center in Baltimore, Maryland. Approval for this study was provided by the institutional review board.

Participant sample

Patient records at the Kennedy Krieger Institute's Department of Neuropsychology were reviewed to identify AYAs aged 14–39 years at the time they underwent clinical neuropsychological testing. These evaluations were completed after clinical referral and included a standard battery of cognitive assessments based on age; scores from the adaptive behavior, intelligence, processing speed, and executive function measures were used for analysis. To provide consistency across the age span, only parent or caregiver ratings were utilized for the adaptive behavior and executive function scales. As such, not all participants had assessments for all measures of interest completed (i.e., when a young adult was not accompanied to the visit by a parent); subsamples with data on each measure of interest are shown in Table 2. Treatment information was confirmed from medical records to verify oncology diagnosis, age at diagnosis, sex, race/ethnicity, and use of CNS-directed radiation therapy.

Table 2.

Adaptive and Neurocognitive Scores Compared to Normative Values

Diagnostic group (n)	Mean	SD	t	Cohen's d
Adaptive function (ABAS)
Adaptive conceptual
All patients (n = 108)	90.30	13.63	−7.399^a	0.677
Leukemia (n = 37)	90.49	14.34	−4.035^a	0.648
Brain tumors (n = 56)	90.38	14.42	−4.996^a	0.654
Solid tumors (n = 15)	89.53	8.58	−4.727^a	0.857
Adaptive social
All patients (n = 108)	90.03	15.23	−6.804^a	0.660
Leukemia (n = 37)	89.97	14.24	−4.283^a	0.686
Brain tumors (n = 56)	90.34	16.21	−4.460^a	0.618
Solid tumors (n = 15)	89.00	14.76	−2.886^b	0.740
Adaptive practical
All patients (n = 108)	87.40	14.15	−9.253^a	0.864
Leukemia (n = 37)	88.57	15.26	−4.558^a	0.756
Brain tumors (n = 56)	86.01	14.44	−7.245^a	0.950
Solid tumors (n = 15)	89.67	9.82	−4.077^a	0.814
Verbal IQ composite
All patients (n = 128)	96.80	17.17	−2.111^b	0.198
Leukemia (n = 40)	95.60	15.52	−1.793	0.288
Brain tumors (n = 71)	96.18	18.93	−1.699	0.224
Solid tumors (n = 17)	102.81	12.16	0.738	0.206
Processing speed
All patients (n = 129)	88.19	14.85	−9.032^a	0.791
Leukemia (n = 40)	90.85	11.69	−4.951^a	0.680
Brain tumors (n = 71)	84.73	16.20	−7.940^a	0.978
Solid tumors (n = 18)	95.94	11.61	−1.482	0.303
Executive function
Metacognitive
All patients (n = 116)	58.46	10.69	8.519^a	0.817
Leukemia (n = 36)	59.92	10.71	5.557^a	0.957
Brain tumors (n = 63)	57.59	10.73	5.586^a	0.732
Solid tumors (n = 17)	58.59	10.61	3.337^a	0.833
Behavior regulation
All patients (n = 116)	50.63	9.83	0.690	0.064
Leukemia (n = 36)	54.19	11.00	2.287^c	0.399
Brain tumors (n = 63)	48.92	9.23	−0.928	0.112
Solid tumors (n = 17)	49.41	7.60	−0.319	0.066

p < 0.001.

p < 0.05.

p < 0.01.

ABAS, Adaptive Behavior Assessment System; IQ, intelligence quotient.

Measures

Adaptive function was measured with the Adaptive Behavior Assessment System (ABAS) second and third editions,^23,24 a psychometrically reliable and validated tool for caregiver report of adaptive behaviors and skills for daily life. The measure evaluates nine adaptive skill areas and provides a norm-referenced composite score for three domains (conceptual, social, and practical) and an overall summary adaptive composite. The conceptual domain assesses behaviors required for communicating with others, managing and accomplishing needed tasks, and applying academic skills to daily tasks. The practical domain measures behaviors needed to manage home, classroom, or work settings, care for personal hygiene, and navigate one's community. Social domain behaviors include interacting interpersonally, acting with social responsibility, and engaging appropriately in leisure activities. Higher scores indicate better adaptive function.

Intelligence was assessed with the age-appropriate standardized measure, including the Wechsler Intelligence Scale for Children, fourth and fifth edition, Wechsler Adult Intelligence Scale fourth edition, or the Reynolds' Intellectual Assessment Scale.^25–29 These measures estimate overall composite intelligence quotient (IQ), verbal and nonverbal reasoning, and/or processing speed. Given the recognized risk of reduced processing speed in survivors, the verbal reasoning composite (Verbal Comprehension Index; Verbal Ability Index) was used as an estimate of cognitive ability, as the verbal composite does not rely upon speeded responding or graphomotor skills, as does the overall or full-scale score.

Processing speed was assessed with the Processing Speed Index (PSI) from the age-appropriate Wechsler measure. The PSI assesses the participant's efficiency of scanning a page of symbols, processing the information, and providing a graphomotor (e.g., pencil and paper) response. All scores are compared against normative age-specific data; higher scores indicate greater processing efficiency.

Executive function was assessed with parent ratings on the Behavior Rating Inventory of Executive Function (BRIEF), first and second editions.^30,31 The BRIEF is a psychometrically sound rating scale designed to assess the application of executive skills in the individual's typical daily environment and yields age-referenced summary behavior regulation and metacognition (e.g., cognitive regulation) composite scores relative to the normative sample. Higher scores indicate greater difficulty with executive function. As with the adaptive ratings, parent ratings rather than self ratings were used to avoid self-report bias or developmental variability in insight.

Diagnosis and treatment information were extracted from medical records. The diagnoses were categorized by disease groups; leukemia/B cell lymphoma, brain tumors, and solid tumors. Solid tumors included patients with a history of Ewing sarcomas (n = 4), rhabdomyosarcoma (n = 2), osteosarcoma (n = 1), Hodgkin lymphoma (n = 2), Langerhans cell histiocytosis (LCH; n = 3), nasopharyngeal carcinoma (n = 2), neuroblastoma (n = 2), desmoplastic round cell tumor (n = 1), nongerm cell testicular tumor (n = 1), and testicular cancers (n = 1) that were treated with systemic, potentially neurotoxic, chemotherapy. Patients with LCH were included in the sample as they also received systemic cytotoxic chemotherapy. All solid tumors (other than brain tumors) were grouped together for analysis as the therapy did not include CNS directed treatments. CNS-directed therapy was defined as any targeted radiation therapy delivered to the brain, including targeted boost of CNS radiation for stem cell transplant conditioning, or intrathecal chemotherapy.

Statistical analysis

One-sample t-tests compared mean AYA survivor scores to population normative data across outcomes of interest. Hierarchical linear regression models examined prediction of adaptive composite scores from intelligence (verbal IQ), processing speed, and executive function (behavior regulation and metacognition composites). Given the well-documented impacts that age at diagnosis and use of CNS-directed radiation therapy have on neurocognition,^32–34 these variables were included in all models.

Results

Of the 139 participants included in this study, 108 had ABAS scores available for analysis, which is the primary outcome of interest for this study; participants with missing ABAS data were retained to analyze mean IQ, processing speed, and executive function scores. Table 1 displays participant characteristics. The average age at the time of diagnosis was 11.5 years (standard deviation [SD] = 5.7 years). Just over half of the patients were male (n = 59; 57.6%) and identified as white (n = 95; 68.3%). The most frequent diagnoses were brain tumor (n = 77, 55.4%) and leukemia/non-Hodgkin lymphoma (n = 43, 30.9%), with the remaining 19 (13.7%) patients treated for a solid tumor.

Table 1.

Patient Demographics

	Mean (SD)
Age at diagnosis (years)	11.5 (5.7)
Gender	n (%)
Female	59 (42.4)
Male	80 (57.6)
Race/ethnicity	n (%)
White	95 (68.3)
Black	27 (19.4)
Hispanic	7 (5)
Asian	7 (5)
Multiracial	3 (2.2)
Diagnosis group	n (%)
Brain tumor	77 (55.4)
Leukemia	43 (30.9)
Solid tumor	19 (13.7)
CNS RT	n (%)
Yes	50 (36)
No	89 (64)

CNS, central nervous system; n, number; RT, radiation therapy; SD, standard deviation.

General neurocognitive functioning

AYA survivors of childhood cancers as a group scored significantly below normative mean of 100 (SD = 15) across all adaptive function domains and processing speed (Table 2). Although group mean scores fell largely within one standard deviation of the normative mean, the number of AYA survivors who scored >1SD below the mean was higher than expected based on normative data (Fig. 1). Based on a normative distribution, no more than 16% of participants would be expected to score worse than 1SD from the mean, but 35% to 45% of these AYA participants scored worse than normative means in adaptive conceptual, adaptive social, adaptive practical, metacognition, and processing speed domains.

FIG. 1.

Percent of the sample performing more than 1SD (e.g., >16%) worse than the normative mean for each adaptive and neurocognitive domain. Bold horizontal line indicates the expected 16% ≤ 1SD of the normative mean. SD, standard deviation.

As a group, their verbal IQ scores also differed significantly from the normative mean, but this difference was not present when evaluated based on diagnosis group; it is important to note that all group means fell within five standard score points of the mean (e.g., 0.33SD) and well within the average range. Likewise, mean processing speed scores were significantly below the normative mean of 100 (SD = 15) for AYA survivors as a group (Cohen's d = 0.791), as well as for the brain tumor (d = 0.978) and leukemia (d = 0.680) groups, but not for patients with a history of solid tumors (mean = 95.94; d = 0.303). With regard to executive functioning, all AYAs were consistently rated as showing significantly higher scores (indicating greater difficulty) relative to age-referenced norm of 50 (SD = 10) for metacognitive executive skills (Cohen's d = 0.817). There was not a significant difference in behavioral regulation skills (d = 0.064) with the exception of AYAs with a history of leukemia, who also exhibited somewhat more difficulty with behavioral regulation than expected for age (d = 0.399).

Adaptive behavior

Utilizing hierarchical regression analysis, we examined the impact of each hypothesized predictor on adaptive function domain while controlling for treatment-related variables of age at diagnosis and use of CNS-directed radiation therapy, neither of which predicted adaptive outcomes (Table 3). Measures of verbal intelligence, processing speed, and executive function (behavior regulation and metacognition composite scores) were entered into the model individually and sequentially to evaluate their contribution to prediction of adaptive behavior scores (e.g., change in R², or ΔR²).

Table 3.

Hierarchical Linear Regression on Adaptive Function Scores

	Cumulative model R²	Block ΔR²	β	p
Analysis 1: adaptive conceptual
Age at diagnosis	0.007		0.086	0.418
CNS radiation	0.007	0.000	0.009	0.935
Verbal IQ	0.094	0.086	0.294	0.005
Processing speed	0.146	0.053	0.242	0.023
Metacognitive	0.333	0.186	−0.441	0.000
Behavior Regulation	0.371	0.039	−0.248	0.026
Analysis 2: adaptive social
Age at diagnosis	0.021		0.147	0.166
CNS radiation	0.022	0.000	−0.010	0.924
Verbal IQ	0.026	0.005	0.070	0.510
Processing speed	0.053	0.026	0.170	0.128
Metacognitive	0.226	0.173	−0.425	0.000
Behavior Regulation	0.261	0.035	−0.237	0.049
Analysis 3: adaptive practical
Age at diagnosis	0.027		0.166	0.117
CNS radiation	0.029	0.001	−0.037	0.733
Verbal IQ	0.083	0.054	0.234	0.026
Processing speed	0.160	0.077	0.292	0.006
Metacognitive	0.266	0.105	−0.331	0.001
Behavior Regulation	0.271	0.005	−0.089	0.452

Bold indicates p-values < 0.05.

Each variable entered sequentially in separate blocks; β = beta for each block sequentially.

Adaptive conceptual

After controlling for the treatment related variables of age at diagnosis and CNS radiation, verbal IQ scores accounted for a significant proportion of variability in parent-rated conceptual composite scores (ΔR² = 0.086, p = 0.005). Over and above the treatment related variables, the addition of processing speed accounted for an additional 5.3% (p = 0.023) of the variance in adaptive conceptual skills. Finally, both the metacognition (ΔR² = 0.186, p < 0.001) and behavior regulation (ΔR² = 0.039, p = 0.026) composites added significantly to prediction of conceptual adaptive functioning. Even with executive function scores in the model, verbal IQ (β = 0.202, p = 0.028) and processing speed (β = 0.281, p = 0.03) remained unique predictors of AYA's conceptual adaptive skills. Taken together, the model accounted for 37.1% of the variability in adaptive skills.

Adaptive social

After controlling for the treatment related variables, neither verbal IQ nor processing speed accounted for a significant amount of variance in adaptive social functioning (VIQ: ΔR² = 0.005, p = 0.510; PSI: ΔR² = 0.026, p = 0.128). However, the metacognition composite accounted for an additional 17.3% of the variance (p < 0.001), with behavior regulation also contributing uniquely (ΔR² = 0.035, p = 0.049) to social adaptive domain scores. With all predictors in the model, processing speed (β = 0.208, p = 0.039) emerged as an additional unique predictor of social functioning. Taken together, the model explained 26.1% of the variance in social adaptive functioning.

Adaptive practical

After controlling for the treatment related variables, verbal IQ accounted for an additional 5.4% of the variance (p = 0.026) in practical adaptive domain scores. With treatment related variables and verbal IQ in the model, processing speed also explained a significant proportion of additional variance (ΔR² = 0.077, p = 0.006). Finally, metacognitive executive skills (ΔR² = 0.105, p = 0.001), but not behavioral regulation (ΔR² = 0.005, p = 0.452), accounted for unique additional variance in practical adaptive functioning. With all predictors in the model, processing speed (β = 0.322, p = 0.002), but not verbal IQ (p = 0.174), remained a significant unique predictor of practical composite scores. Taken together, the model explained 27.1% of the variance in practical adaptive skills.

Discussion

Findings from this cross-sectional study show that AYA survivors of childhood cancers referred for neuropsychological testing exhibit significant weaknesses across cognitive domains. These effects are not limited to patients with a history of brain tumors, leukemia/lymphoma, or CNS-directed therapy; groups commonly considered to be at highest risk. There are important cognitive factors that explain these deficits beyond diagnosis and therapy alone.

Cognitive profile of AYA survivors

Similar to prior studies, our investigation demonstrated that survivors of childhood cancers are at risk for lasting neurocognitive effects.^5,6,8,15,35 The AYA survivors in this study exhibited substantive impacts on all domains of adaptive function when examined as an entire group compared to normative data, with moderate to large effect sizes (Cohen's d > 0.660 for all domains). Survivors also had significantly slower processing speed and metacognitive executive dysfunction compared to normative data. It is notable that, on average, these AYAs demonstrated broadly average intelligence, suggesting that the functional weaknesses observed in adaptive and executive function are not solely due to reduced IQ. With regard to executive function, cognitive aspects of self- and task-regulation (e.g., metacognition) were more consistently affected than behavioral and emotional regulation (large vs. small effects, respectively).

Importantly, these differences in neurocognitive functioning were not limited to patients who received CNS-directed therapy. Patients with a history of solid tumors who did not receive intrathecal chemotherapy or CNS radiation therapy also exhibited weaknesses in adaptive function and metacognitive executive functioning. Oncology providers who care for these patients should be aware that the lack of cranial radiation, brain surgery, and CNS-directed therapy does not prevent these patients from developing neurocognitive deficits and should consider the impacts that systemic chemotherapy or a cancer diagnosis alone may have on key domains of functioning. This study suggests that patients who have a history of solid tumors, who have previously not been thought of as high risk, may develop some neurocognitive challenges after cancer therapy. The small sample size and heterogeneity of diagnoses within this group preclude identification of which types of solid tumor diagnoses might present the most risk; however, these data do suggest the importance of ensuring access to neurocognitive screening and/or assessment for these patients.

Predictors of neurocognitive function and adaptive behavior

Intelligence, processing speed, and executive function together explained a substantial amount of variability in adaptive function in AYA survivors of childhood cancers, over and above that explained by age at diagnosis and CNS-directed therapy. Overall, metacognitive executive function was the most consistent and sizable predictor of adaptive functioning across domains. The scales comprising the metacognition composite assess skills of initiation, working memory, planning and organizing (setting goals, establishing steps to reach goals, carrying out tasks), organization of materials, and self-monitoring.^30,31 These are critical life skills for meeting daily expectations across settings (e.g., home, classroom, and work environments), and it is clear that the lack of these key skills could limit day to day performance substantially, even in the context of generally age-appropriate IQ. As metacognition was most consistently and significantly correlated with adaptive behaviors, identification and implementation of strategies to mitigate impacts to metacognition during and following cancer therapy may have profound impacts on the daily lives of survivors.

Behavior regulation added to the hierarchical model of conceptual and social adaptive functioning for survivors beyond metacognition, speed, and verbal IQ. Behavior regulation includes the ability to control impulses, easily transition from one activity/situation to another, and appropriately regulate emotional responses.^30,31 The ability to regulate one's own behavioral reactions can impact functional adaptive behaviors (e.g., initiating or completing needed tasks even when the individual prefers another activity) and influence an individual's ability to excel in their environment.

Processing speed was significantly associated with adaptive conceptual and adaptive practical functioning for survivors. The inability to quickly ingest, interpret, and use information has a negative impact on the ability to develop adaptive conceptual skills and may impact skills needed to communicate, accomplish tasks, and utilize academic skills efficiently. Likewise, delayed processing speed has some influence on the ability to develop adaptive practical skills for self-sufficiency, self-care, and future-oriented planning tasks.

Predictors of diagnosis and radiation with adaptive function

In contrast to prior work,^5–8 we did not find that age at diagnosis or CNS radiation correlated with functional outcomes. We did find that all diagnosis groups had lower than expected adaptive function scores independent of CNS radiation. Data utilized in this study were extracted from records of patients who were referred for neuropsychological evaluation, meaning that they likely had some cognitive function concerns that prompted a referral from the parent, patient, or oncology team, which introduces some selection bias and may explain why there is not a notable difference due to radiation therapy.

Nonetheless, these findings suggest that neurocognitive challenges after treatment are not limited to those who received CNS-directed therapy. Rather, children and AYAs with solid tumors, those who do not receive radiation therapy, and patients who do not undergo brain surgery are also at risk for adaptive behavior difficulty after therapy. Furthermore, there may be some impact of the lived experience of having cancer or undergoing cancer treatment as a child that limits the development of executive function and adaptive behavior skills because treatment interrupts typical development and can result in fewer opportunities to take on greater independence.

Limitations

Limitations of this study arise mainly from the retrospective cross-sectional design. First, there are no pretreatment neuropsychological assessments available for comparison, removing the ability to determine degree of change after diagnosis and treatment. However, the literature is rich with descriptions of negative neurocognitive impacts from therapy, which suggests that the uniform deficits across domains seen in this study are likely related to diagnosis and treatment. Second, there is a strong selection bias within this sample because all patients for whom these data were extracted were referred for neuropsychological testing and likely exhibited a degree of neurocognitive difficulties that warranted a need for further evaluation and management and, therefore, may not reflect the entirety of the AYA survivor population.

Future directions

Future research should focus on determining the degree to which these patients experience neurocognitive changes after therapy for childhood cancers, including longitudinal prospective evaluations. There is little evidence yet that specific, noncontextual executive function interventions provide lasting improvements in these skills for survivors of childhood cancer. The use of computerized cognitive interventions and telephone-based coaching for children with a history of brain tumor or leukemia has shown some potential for supporting working memory, processing speed, and cognitive flexibility.^36,37 Supporting the use of technology either through web-based intervention programs or online calendars, reminders, and organizational platforms may support development of executive skills or provide a means to mitigate the daily impacts of lower executive skills.³⁸ Structured interventions with cognitive behavioral therapy, cognitive training, relaxation coaching, stress management, and occupational therapy have also been suggested as potentially beneficial for survivors.³⁹ Interventions tailored to the challenges imposed by chemotherapy, radiation therapy, and hospitalization that accompany a cancer diagnosis may prove to be more successful in improving neurocognitive functioning than existing standard approaches.

The Children's Oncology Group notes that survivors who have received CNS radiation, brain surgery, or therapy with cytarabine and/or methotrexate (which mainly includes children with brain tumors or leukemia) are at risk for neurocognitive effects from therapy and should be monitored for deficits after treatment.⁴⁰ However, these findings suggest that patients who did not receive these therapies may still develop cognitive challenges that can impact daily adaptive outcomes.

Current practices of referring survivors for neuropsychological evaluation based on disease or treatment associated factors may miss patients who require neuropsychological services. Survivors of childhood cancers should undergo routine screening for difficulties in domains of adaptive function after therapy during follow-up with the oncology team⁴¹ so that providers identify the need for referral early in the survivorship phase. Standardizing the lexicon used to define the role of neuropsychology teams in caring for these patients will be important in helping identify survivors most in need of support, and neuropsychologists should be incorporated into the interdisciplinary cancer care team.⁴²

Conclusion

This study reinforces prior findings that AYAs who have completed treatment for childhood cancer may have neurocognitive deficits that could impact the ability to fully meet expectations for age appropriate functioning. Adaptive function reflects how well a survivor is able to navigate and function independently in adulthood, and we identify some previously unexplored relationships that suggest weaknesses in intelligence, processing speed, and/or executive functioning after therapy, which may substantially limit day to day adaptive outcomes. By improving understanding of how these deficits manifest, and identifying strategies to support AYA survivors, the burden of cancer treatment in children may be eased so that survivors have the opportunity to reach their fullest functional potential.

Footnotes

Acknowledgments

None to disclose, all persons who had contributed meaningful input to this work are listed as authors.

Author Disclosure Statement

Authors have no disclosures to state; there are no conflicts of interest to declare.

Funding Information

No funding was received for this project.

References

National Cancer Institute. Surveillance, epidemiology, and end results program. Recent trends in survival, all cancer sites combined, ages <15, 15–39. Updated May 2020. Accessed June 3, 2020 from: https://seer.cancer.gov

John

, Sender

, Bota

. Cognitive impairment in survivors of adolescent and early young adult onset non-CNS cancers: does chemotherapy play a role?. J Adolesc Young Adul Oncol. 2016; 5(3):226–31.

Nicklin

, Velikova

, Hulme

, et al. Long-term issues and supportive care needs of adolescent and young adult childhood brain tumour survivors and their caregivers: a systematic review. Psychooncology. 2019; 28(3):477–87.

Pulsifer

, Duncanson

, Grieco

, et al. Cognitive and adaptive outcomes after proton radiation for pediatric patients with brain tumors. Int J Radiat Oncol Biol Phys. 2018; 102(2):391–8.

Netson

, Conklin

, Wu

, et al. Longitudinal investigation of adaptive functioning following conformal irradiation for pediatric craniopharyngioma and low-grade glioma. Int J Radiat Oncol Biol Phys. 2013; 85(5):1301–6.

Papazoglou

, King

, Morris

, Krawiecki

. Parent report of attention problems predicts later adaptive functioning in children with brain tumors. Child Neuropsychol. 2008; 15(1):40–52.

Van Der Plas

, Erdman

, Nieman

, et al. Characterizing neurocognitive late effects in childhood leukemia survivors using a combination of neuropsychological and cognitive neuroscience measures. Child Neuropsychol. 2018; 24(8):999–1014.

Raghubar

, Orobio

, Ris

, et al. Adaptive functioning in pediatric brain tumor survivors: an examination of ethnicity and socioeconomic status. Pediatr Blood Cancer. 2019; 66(9):e27800.

Vetsch

, Wakefield

, McGill

, et al. Educational and vocational goal disruption in adolescent and young adult cancer survivors. Psychooncology. 2018; 27(2):532–8.

10.

Hydeman

, Uwazurike

, Adeyemi

, Beaupin

. Survivorship needs of adolescent and young adult cancer survivors: a concept mapping analysis. J Cancer Surviv. 2019; 13(1):34–42.

11.

Ashford

, Netson

, Clark

, et al. Adaptive functioning of childhood brain tumor survivors following conformal radiation therapy. J Neurooncol. 2014; 118(1):193–9.

12.

Papazoglou

, King

, Morris

, Krawiecki

. Cognitive predictors of adaptive functioning vary according to pediatric brain tumor location. Dev Neuropsychol. 2008; 33(4):505–20.

13.

Husson

, Prins

, Kaal

SEJ

, et al. Adolescent and young adult (AYA) lymphoma survivors report lower health-related quality of life compared to a normative population: results from the PROFILES registry. Acta Oncol. 2017; 56(2):288–94.

14.

Kunin-Batson

, Kadan-Lottick

, Zhu

, et al. Predictors of independent living status in adult survivors of childhood cancer: a report from the childhood cancer survivor study. Pediatr Blood Cancer. 2011; 57(7):1197–203.

15.

Hutchinson

, Pfeiffer

, Wilson

. Cancer-related cognitive impairment in children. Curr Opin Support Palliat Care. 2017; 11(1):70–5.

16.

Kesler

, Ogg

, Reddick

, et al. Brain network connectivity and executive function in long-term survivors of childhood acute lymphoblastic leukemia. Brain Connect. 2018; 8(6):333–42.

17.

Puhr

, Ruud

, Anderson

, et al. Self-reported executive dysfunction, fatigue, and psychological and emotional symptoms in physically well-functioning long-term survivors of pediatric brain tumor. Dev Neuropsychol. 2019; 44(1):88–103.

18.

Papazoglou

, Jacobson

, Zabel

. More than intelligence: distinct cognitive/behavioral clusters linked to adaptive dysfunction in children. J Int Neuropsychol Soc. 2013; 19(2):189–97.

19.

Kanellopoulos

, Andersson

, Zeller

, et al. Neurocognitive outcome in very long-term survivors of childhood acute lymphoblastic leukemia after treatment with chemotherapy only. Pediatr Blood Cancer. 2016; 63(1):133–8.

20.

Kenny

, Cribb

, Pellicano

. Childhood executive function predicts later autistic features and adaptive behavior in young autistic people: a 12-year prospective study. J Abnorm Child Psychol. 2019; 47(6):1089–99.

21.

Pugliese

, Anthony

, Strang

, et al. Longitudinal examination of adaptive behavior in autism spectrum disorders: influence of executive function. J Autism Dev Disord. 2016; 46(2):467–77.

22.

Shultz

, Hoskinson

, Keim

, et al. Adaptive functioning following pediatric traumatic brain injury: relationship to executive function and processing speed. Neuropsychology. 2016; 30(7):830–40.

23.

Oakland

, Harrison

. ABAS-II: clinical use and interpretation. II ed. San Diego: Academic Press;. 2008.

24.

Harrison

, Oakland

Adaptive behavior assessment system (ABAS). 3rd ed. Torrence, CA: Western Psychological Services; 2015.

25.

Wechsler

Wechsler intelligence scale for children. 4th ed. San Antonio, TX: Psychological Corporation;. 2003.

26.

Wechsler

Wechsler intelligence scale for children. 5th ed. San Antonio, TX: NCS Pearson;. 2013.

27.

Wechsler

Wechsler intelligence scale for children. 4th ed. San Antonio, TX: NCS Pearson;. 2008.

28.

Wechsler

Wechsler intelligence scale for children: Fifth edition. Accessed January 30, 2020 from: https://www.pearsonassessments.com/store/usassessments/en/Store/Professional-Assessments/Cognition-%26-Neuro/Gifted-%26-Talented/Wechsler-Intelligence-Scale-for-Children-%7C-Fifth-Edition%2C-Integrated/p/100001322.html

29.

Reynolds

, Kamphaus

Reynolds intellectual assessment scales. Accessed July 08, 2020 from: https://www.parinc.com/Products/Pkey/364

30.

Gioia

, Isquith

, Guy

, Kenworthy

BRIEF: behavior rating inventory of executive function. Lutz, FL: Psychological Assessment Resources, Inc.; 2000.

31.

Gioia

, Isquith

, Guy

, Kenworthy

BRIEF 2: behavior rating inventory of executive function, second edition. Lutz, FL: Psychological Assessment Resources, Inc.; 2015.

32.

Castellino

, Ullrich

, Whelen

, Lange

. Developing interventions for cancer-related cognitive dysfunction in childhood cancer survivors. J Natl Cancer Inst. 2014; 106(8):dju186.

33.

Prasad

, Hardy

, Zhang

, et al. Psychosocial and neurocognitive outcomes in adult survivors of adolescent and early young adult cancer: a report from the childhood cancer survivor study. Am J Clin Oncol. 2015; 33(23):2545–52.

34.

Tonning Olsson

, Perrin

, Lundgren

, et al. Long-term cognitive sequelae after pediatric brain tumor related to medical risk factors, age, and sex. Pediatr Neurol. 2014; 51(4):515–21.

35.

Heitzer

, Ashford

, Hastings

, et al. Neuropsychological outcomes of patients with low-grade glioma diagnosed during the first year of life. J Neurooncol. 2019; 141(2):413–20.

36.

Conklin

, Ogg

, Ashford

, et al. Computerized cognitive training for amelioration of cognitive late effects among childhood cancer survivors: a randomized controlled trial. J Clin Oncol. 2015; 33(33):3894–902.

37.

Kesler

, Lacayo

, Jo

. A pilot study of an online cognitive rehabilitation program for executive function skills in children with cancer-related brain injury. Brain Injury. 2011; 25(1):101–12.

38.

National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology: Survivorship. NCCN.org Web site. Updated 2019. Accessed February 10, 2020 from: https://www.nccn.org/professionals/physician_gls/default.aspx

39.

National Cancer Institute. Adolescents and young adults with cancer. Updated 2018. Accessed January 20, 2020 from: https://www.cancer.gov/types/aya

40.

Children's Oncology Group. Long-term follow-up guidelines for survivors of childhood, adolescent, and young adult cancers. 5th ed. Arcadia, CA: Children's Oncology Group;. 2018.

41.

Annett

, Patel

, Phipps

. Monitoring and assessment of neuropsychological outcomes as a standard of care in pediatric oncology. Pediatr Blood Cancer. 2015; 62:S460–513.

42.

Baum

, Powell

, Jacobson

, et al. Implementing guidelines: proposed definitions of neuropsychology services in pediatric oncology. Pediatr Blood Cancer. 2017; 64(8):e26446.