Abstract
Central nervous system (CNS)–directed administration chemotherapy for the treatment of acute lymphoblastic leukemia (ALL) has led to survival rates in excess of 90%; however, late but damaging effects on white matter tracts in cortical and subcortical regions have been identified (Khong et al., 2003). White matter tracts are integral to neural mapping underlying skill acquisition and consolidation (Roberts et al., 2013). Myelination of axons begins in infancy and continues into adulthood (Askins & Moore, 2008). This ongoing neural maturation supports the acquisition of higher-level learning processes beyond the foundational skills learned and developed throughout early childhood (Roberts et al., 2013). Disruption to the maturational process of white matter as a result of CNS-directed therapy in children and adult survivors of ALL suggests the potential for neurocognitive deficits in skills (Anderson et al., 2011; Krull et al., 2013). White matter maturation patterns in adolescence have been linked to developments in cognitive and language function (Schmithorst & Yuan, 2010), and thus neural damage sustained in childhood secondary to intrathecal chemotherapy (ITC) for ALL may manifest in deficits in these functions in later childhood and/or adolescence.
Foundational language skills emerge early in childhood and continue to develop into adolescence; higher-order, more cognitively demanding language forms may, however, not be acquired until early adolescence, and mastery of these skills may not be completed until adulthood. Higher-order language skills enable interpretation and expression of non-literal or abstract meaning (Nippold, 2007). This ability coincides with cognitive maturation and the development of executive functions, enabling coordination of cognitive resources to complete complex tasks, many of which are recognized as underpinning the emergence of more complex early adolescent language (Ciccia et al., 2009; Nippold, 2014). The development of abstract and figurative language continues through adolescence into adulthood (Nippold, 2007). Of concern, these higher-order language skills have been found to be compromised in adolescents with a history of white matter trauma (Moran & Gillon, 2005; Yang et al., 2010). Because higher-order language is associated with emotional regulation (Lindsay & Dockrell, 2012), social acceptance among peers (Askins & Moore, 2008), academic success, and quality of life (Kaiser & Roberts, 2011), there is a need to explore the language abilities of children who receive ITC for ALL, especially as language deficits may contribute to the reported activity and participation restrictions experienced by adult survivors of childhood leukemia (Stuber et al., 2010). Investigations into the language effects following the administration of prophylactic ITC for childhood ALL have reported deficits in expressive language and figurative language (Lewis et al., 2011).
The aim of this study was to investigate the language abilities in children/adolescents with a childhood history of ITC for ALL older than the children investigated by Dowling et al. (2014). As previous research has suggested that abilities consolidated prior to treatment may remain intact (Didus et al., 1999), the study, while including assessment of foundational language skills, focused on higher-order language skills in five children at the late childhood/adolescent stage of language development. This study examined language skills in older children/adolescents with a childhood history of ITC for ALL.
Method
Five participants (four females and one male) between 9 and 18 years of age with a history of a diagnosis of ALL and treatment with ITC participated in the study. Age at diagnosis (and treatment initiation) ranged from 2 years 7 months to 11 years 10 months.
Each child’s foundational skills were assessed. Receptive vocabulary was assessed with the Peabody Picture Vocabulary Test–Third Edition (PPVT-3; Dunn & Dunn, 1997), whereas broader foundational language skills were assessed using the Clinical Evaluation of Language Fundamentals–Fourth Edition (CELF-4; Semel et al., 2003). Later-developing higher-order language skills that emerge in late childhood/early adolescence, such as using and understanding inferential, metaphoric, and figurative language, were assessed using the Test of Language Competence–Expanded Edition (TLC-E; Wiig & Secord, 1989), which assesses emerging metalinguistic competence through four tests of non-literal language use: Ambiguous Sentences (AS), Listening Comprehension: Making Inferences (LCMI), Oral Expression: Recreating Sentences (OERS), and Figurative Language (FL). The AS subtest requires the adolescent to provide two alternate interpretations of sentences that offer ambiguity at the word or sentence level. The LCMI subtest requires the adolescent to provide plausible inferences based on lead-in and lead-out casual events. In the OERS subtest, the adolescent is required to use three given words to construct sentences that are grammatically correct and that are appropriate to the constraints of a communicative situation presented in the picture prompt. For the FL subtest, the adolescent is required to interpret figurative expressions such as idioms and metaphors. The Test of Problem Solving (TOPS; Bowers et al., 2005, 2007) was administered for assessment of higher-order social problem solving.
Results
There was no consistent language profile in the data based on age at language assessment or time post-diagnosis. Two participants had intact foundational and higher-order language skills. Two participants presented with intact foundational language skills but difficulties with higher-order language tasks. One participant was identified as having clinically significant difficulty on both foundational and higher-order language skills. Although her foundational language results present borderline clinical concern only, she presented with deficits of clinical concern across all eight assessed higher-order adolescent language skill areas that reflect competence in using or understanding non-literal language forms. Certain subtests were implicated more than others across the participants’ profiles. Of the eight higher-order language areas assessed, performance on the TOPS, Expressing Intents Composite (EIC), and TLC-E was compromised to a clinically significant degree for at least two of the five participants, whereas performance on the OERS subtest of the TLC-E was identified as impaired to a clinically significant degree for three of the five participants involved in the study.
Discussion
Although preliminary, these findings, based on the language skills of a small number of participants up to 8 years post-diagnosis, offer some support for the hypothesis that, following ITC-induced neural injury, CNS neurotoxicity may subsequently affect later-developing, more complex language skills that develop in tandem with the continuation of myelination into adulthood. Aarsen et al. (2009) suggested that deficits following neural trauma may emerge as the cognitive demands of tasks increase as the child matures. The findings of the current study offer tentative support for this view. All but one child had age-appropriate foundational language skills, yet three of the five presented with clinically deficient higher-order language skills and each of these three participants experienced particular difficulty on the sentence recreation task (OERS subtest). The EIC from the TLC-E, formed by the constituent subtests of ambiguous sentence explanation and sentence recreation, also offers insight into higher-order expressive language skills of the study’s participants. Two of the five children presented with clinically significant deficits on the EIC, a composite score that reflects their ability to verbally convey complex or high-level material. Variable skill levels on higher-order tasks marked the individual presentations. Participant 2 presented with skill strengths ranging from borderline on figurative language to severe deficits on the TLC-E’s (Wiig & Secord, 1989) recreating novel sentences (OERS subtest). Likewise, Participant 3 presented with borderline overall performance on the TLC-E but severe deficits on one of the subtests (OERS [recreating novel sentences]) that contributes to the overall TLC-E score. Participant 5 also presented with contrasting performance abilities, whereby her ability to recreate novel sentences was moderate-to-severely deficient while all other performance scores indicated intact skill development. Two of the five participants experienced clinical deficits in social problem solving, as reflected in the TOPS (Bowers et al., 2005, 2007)