Abstract
Background:
The impact of growth hormone deficiency (GHD) on neurocognitive function is poorly understood in survivors of childhood acute lymphoblastic leukemia (ALL). This study examined the contribution of GHD to functional outcomes while adjusting for cranial radiation therapy (CRT).
Methods:
Adult survivors of ALL (N=571 [49% female]; mean age=37.4 years, range: 19.4–62.2) completed neurocognitive tests and self-reported neurocognitive symptoms, emotional distress and quality of life. GHD was defined as previous diagnosis of GHD or plasma IGF-1 < −2.0 standard deviations (SD) for sex and age at the time of neurocognitive testing. Hypothyroidism, hypogonadism, sex, age at diagnosis, dose of CRT, and intrathecal and high-dose intravenous methotrexate were included as covariates in multivariable linear regression models.
Results:
Of the 571 survivors, 298 (52%) had GHD; those with GHD were treated with higher CRT dose (p=0.002). Survivors with GHD, irrespective of prior growth hormone treatment, demonstrated poorer vocabulary (z = −0.84 versus −0.61, p=0.02), processing speed (z = −0.49 versus −0.30, p=0.04), cognitive flexibility (z = −1.37 versus −0.94, p=0.01), and verbal fluency (z = −0.74 versus −0.44, p=0.001), and self-reported more neurocognitive problems and poorer quality of life compared to survivors without GHD. Multivariable and mediation models revealed GHD was associated with small effects on quality of life (general health p=0.01, vitality p=0.01, mental health p=0.01); CRT dose accounted for the lower neurocognitive outcomes.
Conclusions:
Adult survivors of childhood ALL treated with CRT are at risk for GHD, though poor neurocognitive outcomes are determined by CRT dose not GHD.
Keywords: growth hormone deficiency, neurocognitive, leukemia survivors
Precis:
Adult survivors of childhood ALL treated with cranial radiation therapy (CRT) are at risk for growth hormone deficiency (GHD). However, poor neurocognitive outcomes are determined by CRT dose not GHD.
Introduction
Long-term survivors of childhood acute lymphoblastic leukemia (ALL) are at risk for neurocognitive problems, which can emerge years after diagnosis and may be progressive with time.(1) Such problems are recognized late effects of cranial radiation therapy (CRT), intrathecal methotrexate, and high dose intravenous methotrexate.
Survivors treated with CRT are also at risk for neuroendocrine problems, particularly growth hormone deficiency (GHD).(2) In a recent report from the SJLIFE Cohort, among survivors treated with CRT, GHD was the most common endocrinopathy, identified in 336 (46.5%) of 748 survivors with an estimated cumulative incidence of 72.4% at 40 years from diagnosis. Further, 78% of those survivors (n=262) with GHD had no other endocrinopathies.
GHD has been associated with altered brain development and neurocognitive problems in the general population. Growth hormone alters neurogenesis, myelin synthesis and dendritic branching.(3) During childhood, GHD has been associated with lower grey matter volumes in basal ganglia, thalamus and hippocampal brain structures, as well as lower white matter integrity (as reflected through fractional anisotropy on diffusion tensor imaging), and memory deficits.(4, 5) Among 43 adults treated with radiation for nasopharyngeal cancer, 95% developed growth hormone deficiency and they reported poor emotional and cognitive quality of life at follow-up.(6) Among 13 adult survivors of childhood ALL treated with CRT, memory problems were observed to improve along with increase in plasma IGF-1 levels following growth hormone treatment.(7)
The contribution of GHD to known effects of CRT on neurocognitive function in long-term survivors of childhood cancer has not been carefully examined. A brain that is already compromised by CRT may be more sensitive to the impact of GHD on neurocognitive function. If a unique contribution from GHD exists, intervention with growth hormone to improve neurocognitive function could be considered. Some survivors are being placed on growth hormone to improve function, and at least one registered clinical trial is examining growth hormone replacement to improve quality of life. We examined the potential contribution of GHD, both treated and untreated, to neurocognitive outcomes and quality of life in long-term survivors of childhood ALL, controlling for the confounding effects of CRT, as well as intrathecal and high-dose intravenous methotrexate.
Methods
The institutional review board at St. Jude Children’s Research Hospital approved this study.
Participants
Participants included survivors of ALL enrolled in the St. Jude Lifetime Cohort (SJLIFE) study, which has been previously described.(8, 9) SJLIFE is a retrospectively identified cohort with longitudinal follow-up. Survivors eligible for this analysis received cranial irradiation as part of their anticancer therapy, had follow-up at St. Jude Children’s Research Hospital, and were at least 18 years of age and 10 or more years from diagnosis at study evaluation. Survivors were excluded if they had history of a neurologic condition or genetic disorder associated with neurocognitive impairment, but unrelated to cancer diagnosis or treatment. We identified 811 survivors eligible for participation, 240 of whom were non-participants, leaving 571 (70.4%) available for the current analyses.
Procedures
Medical records were abstracted to obtain information on CRT exposure and cumulative doses of intrathecal and high-dose intravenous methotrexate. History of GHD was defined by known diagnosis of GHD or plasma IGF-1 < −2.0 z-score for sex and chronological age. Hypothyroidism was defined by previous diagnosis and/or treated central or primary hypothyroidism or presence of Free T4 level < 0.9 ng/dL (12 pmol/L). Hypogonadism was defined by previous diagnosis and/or treated central hypogonadism, premature ovarian insufficiency (females) or Leydig cell failure (males). New diagnoses of hypogonadism were based on estradiol levels <17 pg/mL (0.6 pmol/L) in amenorrheic women <40 years and total testosterone levels <200 ng/dL (7 nmol/L) in males.(2, 10)
The primary outcome of interest was neurocognitive function assessed by standardized testing, including measures of attention (sustained [Continuous Performance Test](11), span [Digit Span forward](12), focused [Trail Making Part A](13)), memory (verbal new learning and delayed recall [California Verbal Learning Test](14), visual [Test of Memory and Learning]), executive function (flexibility [Trail Making Part B](13), fluency [Controlled Oral Word Association Test](13), working memory [Digit Span backward](12)), processing speed (visual [Symbol Search](12), motor [Grooved Pegboard](13), visuomotor [Digit Symbol](12)), intelligence (verbal and perceptual [Wechsler Abbreviated Scale of Intelligence](15)), and academics (reading and mathematics [Woodcock-Johnson](16)). Secondary outcomes included patient-reported neurobehavioral problems (task efficiency, memory, organization, emotional regulation [CCSS-Neurocognitive Questionnaire](17)), emotional distress (depression, anxiety, somatization [Behavior Symptom Inventory-18](18)), and health related quality of life (physical function, role limitations – physical, bodily pain, general health, vitality, social function, role limitations – emotional, mental health [Short-form 36](19)).
Statistical analyses
Demographic and treatment characteristics of participants with or without GHD were compared using Chi-square or t-test, as appropriate. Age adjusted z-scores (μ=0, σ=1.0) were calculated for each neurocognitive and patient-reported outcome, with lower scores reflecting worse performance/symptoms across all measures. These scores were compared between survivors with or without GHD using two-sample t-test. Those scores that demonstrated group differences were carried forward to multivariable modeling. The associations between GHD and these outcomes were evaluated using generalized linear models, adjusted for sex, age at diagnosis, time since diagnosis, cumulative CRT dose, cumulative intrathecal methotrexate (either single or when combined with cytarabine and hydrocortisone) and high-dose intravenous methotrexate. The presence of central hypothyroidism and central hypogonadism, which often co-occur following cranial irradiation, were also included in these models. Multivariable models were also examined to compare the subset of survivors who had GHD without hypothyroidism or hypogonadism (n=172) to survivors without GHD, hypothyroidism or hypogonadism (n=196). Scores that were associated with GHD in univariate analyses were examined in mediation analyses to determine the direct effect of CRT on functional outcomes, and indirect effect of CRT through GHD. To explore associations between age at onset of GHD and the impact of prior growth hormone replacement therapy with neurocognitive outcomes, we compared survivors with childhood onset versus adult onset (i.e. diagnosed at follow-up participation in the SJLIFE study). No survivors in the current study were on growth hormone replacement at the time of evaluation. Univariate comparisons were conducted using t-tests, and multivariable models were not conducted due to the relatively small sample size of childhood onset and prior replacement therapy in these ALL survivors.
Results
Of the 571 participants, 298 (52%) had GHD, 97 (17%) hypothyroidism, and 150 (26%) hypogonadism. No significant differences were observed between survivors with or without GHD by sex (52% v. 45% female, p=0.08), age at testing (37.5 v. 37.3 years, p=0.80), or in any area of social attainment (i.e. education, employment, income, insurance status, marital status, independent living) (Table 1). Compared to survivors without GHD, those with GHD were younger at diagnosis (7.1 v. 5.5 years, p<0.001) and received higher doses of CRT (20.9 v. 22.1 Gy, p=0.002).
Table 1.
Survivor demographic and clinical characteristics.
| Non-GHD (N=273) | GHD (N=298) | P-value | ||
|---|---|---|---|---|
| n (%) | n (%) | |||
| Sex | Female | 122 (44.7) | 155 (52.0) | 0.08 |
| Male | 151 (55.3) | 143 (48.0) | ||
| Race | White | 243 (89.0) | 277 (93.0) | 0.10 |
| Non-White | 30 (11.0) | 21 (7.0) | ||
| Education | High school or less | 86 (33.7) | 96 (33.6) | 0.29 |
| Some college | 74 (29.0) | 99 (34.6) | ||
| College graduate | 95 (37.3) | 91 (31.8) | ||
| Employment | Full-time/student | 173 (68.4) | 183 (64.0) | 0.28 |
| Less than full-time | 80 (31.6) | 103 (36.0) | ||
| Income | <$40,000 | 104 (46.0) | 135 (54.2) | 0.07 |
| ≥$40,000 | 122 (54.0) | 114 (45.8) | ||
| Living situation | Independent | 179 (75.5) | 210 (73.4) | 0.58 |
| Dependent | 58 (24.5) | 76 (26.6) | ||
| Marital status | Ever married | 179 (68.3) | 186 (63.3) | 0.21 |
| Never married | 83 (31.7) | 108 (36.7) | ||
| Insurance | Insured | 202 (77.4) | 236 (79.4) | 0.55 |
| Uninsured | 59 (22.6) | 61 (20.5) | ||
| Mean (min-max) | Mean (min-max) | |||
| Age at diagnosis | Years | 7.1 (0.2–19.5) | 5.5 (0.2–18.8) | <0.001 |
| Age at evaluation | Years | 37.3 (19.4–62.2) | 37.5 (19.4–61.6) | 0.80 |
| Time since diagnosis | Years | 30.2 (11.3–50.8) | 32.0 (13.5–50.0) | 0.006 |
| Cranial Radiation | Cumulative Gy | 20.9 (5.0–59.4) | 22.1 (10.8–59.4) | 0.002 |
| IT Methotrexate | Cumulative ml | 156.5 (0–552.7) | 174.9 (0–869.4) | 0.10 |
| HDIV Methotrexate | Cumulative G/M2 | 4.0 (0–41.3) | 3.0 (0–30.9) | 0.08 |
IT methotrexate expressed in cumulative milliliters (ml), administered as either single agent or when combined with cytarabine and hydrocortisone.
High-dose intravenous (HDIV) methotrexate expressed in grams per square meter (G/M2)
Survivors with GHD demonstrated lower vocabulary (−0.84 v. −0.61, p=0.020), visual processing speed (−0.49 v. −0.30, p=0.04), flexibility (−1.37 v. −0.94, p=0.01) and fluency (−0.74 v. −0.44. p=0.001) compared to survivors without GHD (Table 2). Survivors with GHD self-reported more problems with task efficiency (p=0.04) and emotional regulation (p=0.02), and reported poorer quality of life in physical functioning (p=0.02), bodily pain (p=0.04), general health (p=0.005), vitality (p=0.005), social function (p=0.03) and mental health (p=0.01) compared to survivors without GHD.
Table 2.
Univariate comparison of neurocognitive, emotional and quality of life outcomes between survivors with and without growth hormone deficiency (GHD).
| Non-GHD | GHD | P-value | |
|---|---|---|---|
| Mean (95% CL) | Mean (95% CL) | Mean | |
| Neurocognitive – Direct Testing | |||
| Intelligence/Achievement | |||
| Vocabulary | −0.61 (−0.75, −0.46) | −0.84 (−0.97, −0.70) | 0.02 |
| Perceptual Reasoning | −0.11 (−0.24, 0.02) | −0.24 (−0.37, −0.11) | 0.15 |
| Reading | −0.45 (−0.54, −0.36) | −0.57 (−0.66, −0.47) | 0.09 |
| Mathematics | −0.76 (−0.91, −0.61) | −0.93 (−1.07, −0.79) | 0.10 |
| Attention | |||
| Focus | −0.37 (−0.57, −0.18) | −0.56 (−0.75, −0.36) | 0.19 |
| Sustain | −0.77 (−1.12, −0.42) | −1.13 (−1.54, −0.72) | 0.19 |
| Variability | −0.50 (−0.66, −0.34) | −0.58 (−0.75, −0.42) | 0.46 |
| Memory | |||
| Span | −0.42 (−0.55, −0.30) | −0.54 (−0.66, −0.42) | 0.17 |
| New Learning | −0.24 (−0.41, −0.08) | −0.28 (−0.43, −0.13) | 0.73 |
| Short-term Recall | −0.22 (−0.38, −0.06) | −0.25 (−0.39, −0.11) | 0.79 |
| Long-term Recall | −0.29 (−0.45, −0.12) | −0.36 (−0.51, −0.21) | 0.53 |
| Processing Speed | |||
| Motor | −0.99 (−1.18, −0.80) | −1.03 (−1.20, −0.86) | 0.78 |
| Visual | −0.30 (−0.42, −0.18) | −0.49 (−0.61, −0.36) | 0.04 |
| Visual-Motor | −0.57 (−0.68, −0.45) | −0.67 (−0.78, −0.56) | 0.22 |
| Executive Function | |||
| Flexibility | −0.94 (−1.17, −0.71) | −1.37 (−1.60, −1.13) | 0.01 |
| Fluency | −0.44 (−0.57, −0.30) | −0.74 (−0.87, −0.62) | 0.001 |
| Working Memory | −0.40 (−0.51, −0.30) | −0.51 (−0.62, −0.41) | 0.16 |
| Neurocognitive – Self-report | |||
| Task Efficiency | −0.93 (−0.75, −1.10) | −1.19 (−1.01, −1.37) | 0.04 |
| Memory | −1.08 (−0.92, −1.23) | −1.25 (−1.10, −1.41) | 0.11 |
| Organization | −0.21 (−0.08, −0.34) | −0.34 (−0.20, −0.48) | 0.19 |
| Emotional Regulation | −0.22 (−0.09, −0.35) | −0.46 (−0.32, −0.60) | 0.02 |
| Emotional Distress | |||
| Depression | −0.04 (−0.17, 0.09) | −0.18 (−0.31, −0.05) | 0.13 |
| Anxiety | 0.18 (0.05, 0.30) | 0.05 (−0.08, 0.18) | 0.17 |
| Somatization | −0.12 (−0.24, −0.01) | −0.32 (−0.44, −0.20) | 0.02 |
| Quality of Life | |||
| Physical Functioning | 0.01 (−0.11, 0.13) | −0.20 (−0.33, −0.08) | 0.02 |
| Role Physical | 0.05 (−0.08, 0.18) | −0.11 (−0.24, 0.02) | 0.09 |
| Bodily Pain | 0.09 (−0.04, 0.21) | −0.10 (−0.23, 0.02) | 0.04 |
| General Health | −0.21 (−0.34, −0.08) | −0.49 (−0.64, −0.35) | 0.005 |
| Vitality | −0.07 (−0.20, 0.06) | −0.33 (−0.46, −0.20) | 0.005 |
| Social Function | −0.12 (−0.23, 0.001) | −0.31 (−0.45, −0.18) | 0.03 |
| Role Emotional | −0.09 (−0.23, 0.04) | −0.25 (−0.39, −0.11) | 0.10 |
| Mental Health | −0.09 (−0.22, 0.04) | −0.34 (−0.47, −0.20) | 0.01 |
Note: All means presented as age-adjusted z-scores (μ = 0, σ = 1.0), lower z-scores reflect worse performance or more symptoms.
In multivariable regression models (Table 3), adjusting for CRT, intrathecal and high-dose intravenous methotrexate, sex, age at diagnosis, and time since diagnosis, GHD was associated with a 0.19 lower z-score for fluency (p=0.046). No other directly tested or self-reported neurocognitive measures were associated with GHD, though CRT was significantly associated with all directly tested neurocognitive measures. For quality of life, GHD was associated with a 0.26 lower z-score in general health (p=0.01), a 0.24 lower z-score in vitality (p=0.01), and a 0.26 lower z-score in mental health (p=0.01). Hypothyroidism was associated with a 0.51 lower score in general health (p<0.001) and a 0.43 lower score in vitality (p=0.001). In multivariable models among survivors with versus without GHD, who did not have hypothyroidism or hypogonadism, GHD was not associated with impaired neurocognitive function (Supplemental Table 1) but was associated with more self-reported emotional regulation problems (0.25 lower z-score, p=0.044) and lower quality of life in social function (0.23 lower z-score, p=0.038) and mental health (0.28 lower z-score, p=0.021).
Table 3.
Multiple regression analyses of neurocognitive and quality of life outcomes that differed between survivors with versus without growth hormone deficiency.
| Vocabulary | Visual Speed | Flexibility | Fluency | Task Efficiency | Emotional Regulation | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Est. | P-value | Est. | P-value | Est. | P-value | Est. | P-value | Est. | P-value | Est. | P-value | ||
| GHD | Yes v. No | −0.08 | 0.41 | −0.07 | 0.45 | −0.21 | 0.21 | −0.19 | 0.046 | −0.16 | 0.20 | −0.19 | 0.06 |
| Hypothyroidism | Yes v. No | −0.06 | 0.67 | −0.16 | 0.17 | −0.28 | 0.23 | −0.25 | 0.051 | −0.37 | 0.03 | 0.01 | 0.98 |
| Hypogonadism | Yes v. No | 0.11 | 0.37 | 0.06 | 0.58 | 0.06 | 0.76 | 0.03 | 0.77 | 0.14 | 0.37 | 0.22 | 0.07 |
| Sex | Female | −0.23 | 0.02 | 0.10 | 0.28 | −0.02 | 0.91 | 0.06 | 0.51 | −0.32 | 0.01 | −0.44 | <0.001 |
| Age at diagnosis | Per year | 0.05 | <0.001 | 0.03 | 0.003 | 0.09 | <0.001 | 0.04 | 0.002 | 0.02 | 0.27 | 0.01 | 0.37 |
| Time since diagnosis | Per 10 years | 0.09 | 0.38 | 0.21 | 0.02 | 0.48 | 0.005 | 0.07 | 0.49 | 0.21 | 0.12 | 0.10 | 0.34 |
| CRT dose | Per 10 Gy | −0.37 | 0.002 | −0.58 | <0.001 | −1.12 | <0.001 | −0.28 | 0.01 | −0.26 | 0.11 | −0.09 | 0.46 |
| IT Methotrexate | Per 100 ml | −0.02 | 0.68 | −0.03 | 0.40 | 0.08 | 0.32 | 0.01 | 0.83 | 0.04 | 0.46 | 0.04 | 0.43 |
| HDIV Methotrexate | Yes v. No | 0.31 | 0.04 | 0.35 | 0.009 | 0.14 | 0.58 | 0.37 | 0.009 | 0.53 | 0.005 | 0.27 | 0.08 |
| Physical Function | Bodily Pain | General Health | Vitality | Social Function | Mental Health | ||||||||
| Est. | P-value | Est. | P-value | Est. | P-value | Est. | P-value | Est. | P-value | Est. | P-value | ||
| GHD | Yes v. No | −0.14 | 0.12 | −0.18 | 0.06 | −0.26 | 0.01 | −0.24 | 0.01 | −0.18 | 0.06 | −0.26 | 0.01 |
| Hypothyroidism | Yes v. No | −0.20 | 0.09 | −0.26 | 0.04 | −0.51 | <0.001 | −0.43 | 0.001 | −0.26 | 0.04 | −0.15 | 0.29 |
| Hypogonadism | Yes v. No | −0.13 | 0.20 | 0.07 | 0.52 | 0.02 | 0.86 | 0.01 | 0.93 | −0.01 | 0.98 | −0.02 | 0.89 |
| Sex | Female | −0.26 | 0.004 | −0.18 | 0.06 | −0.16 | 0.11 | −0.31 | 0.001 | −0.11 | 0.25 | −0.18 | 0.08 |
| Age at diagnosis | Per year | −0.01 | 0.24 | −0.03 | 0.03 | −0.03 | 0.01 | −0.03 | 0.005 | −0.02 | 0.12 | −0.03 | 0.02 |
| Time since diagnosis | Per 10 years | 0.08 | 0.37 | 0.06 | 0.55 | 0.18 | 0.08 | 0.12 | 0.21 | 0.13 | 0.17 | 0.15 | 0.14 |
| CRT dose | Per 10 Gy | −0.46 | <0.001 | −0.21 | 0.046 | −0.08 | 0.52 | −0.12 | 0.29 | −0.23 | 0.03 | −0.06 | 0.58 |
| IT Methotrexate | Per 100 ml | 0.06 | 0.15 | 0.03 | 0.52 | 0.01 | 0.94 | 0.01 | 0.91 | 0.02 | 0.65 | −0.01 | 0.91 |
| HDIV Methotrexate | Yes v. No | 0.08 | 0.56 | 0.05 | 0.75 | 0.49 | 0.001 | 0.20 | 0.15 | 0.26 | 0.06 | 0.24 | 0.11 |
Abbreviations: GHD = growth hormone deficiency, CRT = cranial radiation therapy, IT = intrathecal (either single methotrexate or methotrexate combined with cytarabine and hydrocortisone), HDIV = high-dose intravenous.
Given the known associations between CRT and neurocognitive function, and CRT and GHD, mediation analyses were conducted to determine whether GHD was mediating the effect of CRT on functional outcomes (Table 4). Impairments in neurocognitive outcomes were primarily associated with direct effects of CRT, not indirect effects of CRT through GHD. Contributions from GHD were only observed for general health, vitality and mental health components of quality of life.
Table 4.
Mediation analyses of the effect of cranial radiation (per 10Gy) on functional outcomes mediated by growth hormone deficiency (yes/no).
| Outcome | Direct Effect of Cranial Radiation (95% CI) | Direct Effect P-value | Indirect Effect of Growth Hormone Deficiency (95% CI) | Indirect Effect P-value | Total Effect (95% CI) | Total Effect P-value |
|---|---|---|---|---|---|---|
| Vocabulary | −0.41 (−0.61, −0.20) | <0.001 | −0.01 (−0.04, 0.02) | 0.45 | −0.42 (−0.63, −0.21) | <0.001 |
| Visual Speed | −0.57 (−0.75, −0.38) | <0.001 | −0.01 (−0.03, 0.01) | 0.46 | −0.58 (−0.76, −0.39) | <0.001 |
| Flexibility | −0.93 (−1.28, −0.57) | <0.001 | −0.03 (−0.07, 0.02) | 0.26 | −0.95 (−1.31, −0.59) | <0.001 |
| Fluency | −0.38 (−0.58, −0.19) | <0.001 | −0.03 (−0.05, −0.00) | 0.048 | −0.41 (−0.62, −0.20) | <0.001 |
| Task Efficiency | 0.37 (0.08, 0.65) | 0.011 | 0.02 (−0.01, 0.06) | 0.21 | 0.39 (0.10, 0.68) | 0.008 |
| Emotional Regulation | 0.12 (−0.10, 0.34) | 0.28 | 0.02 (−0.00, 0.05) | 0.100 | 0.14 (−0.09, 0.38) | 0.22 |
| Physical Function | −0.50 (−0.68, −0.31) | <0.001 | −0.02 (−0.04, 0.00) | 0.097 | −0.52 (−0.71, −0.32) | <0.001 |
| Bodily Pain | −0.22 (−0.41, −0.03) | 0.025 | −0.02 (−0.05, 0.00) | 0.053 | −0.24 (−0.45, −0.04) | 0.020 |
| General Health | −0.17 (−0.39, 0.04) | 0.12 | −0.04 (−0.07, −0.01) | 0.011 | −0.21 (−0.45, 0.03) | 0.092 |
| Vitality | −0.15 (−0.34, 0.05) | 0.15 | −0.03 (−0.06, −0.01) | 0.011 | −0.18 (−0.40, 0.04) | 0.11 |
| Social Function | −0.27 (−0.47, −0.08) | 0.005 | −0.02 (−0.05, 0.00) | 0.063 | −0.30 (−0.50, −0.09) | 0.004 |
| Mental Health | −0.07 (−0.28, 0.14) | 0.52 | −0.03 (−0.06, −0.01) | 0.014 | −0.10 (−0.34, 0.13) | 0.39 |
Note: Effects and 95% confidence intervals (CI) presented in age-adjusted z-scores (μ = 0, σ = 1.0), lower z-scores reflect worse performance or more symptoms.
In exploratory analyses (Supplemental Tables 2 and 3), neither age at GHD diagnosis nor history of growth hormone replacement therapy was associated with neurocognitive function (as assessed by direct testing). Compared to survivors diagnosed with GHD during adulthood (n=237), those diagnosed during childhood (n=57) self-reported fewer problems with memory (−1.37 v. −0.75, p=0.002), organization (−0.45 v. 0.15, p=0.001) and emotional regulation (−0.54 v. −0.13, p=0.03). The 51 survivors who had previously received growth hormone replacement therapy self-reported fewer problems with memory (−0.76 v. −1.16, p=0.003) and organization (0.25 v. −0.46, p<0.001) compared to the 243 survivors with no prior therapy.
Discussion
Although GHD was associated with lower self-reported quality of life, it had limited impact on directly assessed or self-reported neurocognitive function. Survivors with GHD were younger at diagnosis and further from diagnosis and were treated with higher doses of CRT. Consistent with current literature on neurocognitive outcomes in survivors of childhood ALL,(1) younger age at diagnosis, longer time since diagnosis, and higher dose of CRT accounted for the poorer neurocognitive outcomes. After adjusting for these established predictors, GHD was associated with a slightly lower score in verbal fluency of roughly one-fifth of a standard deviation. Comparatively, the impact of each 10 Gy dose increase of CRT on objectively tested neurocognitive function in survivors ranged from nearly one-third to over 1.0 standard deviation worse performance. Since CRT and GHD could theoretically be on the same pathway to neurocognitive problems, mediation analyses were conducted to examine direct effects of CRT versus indirect effects of CRT through GHD. Virtually all neurocognitive outcomes were significantly impacted by the direct effect of CRT rather than the indirect effects of CRT through GHD.
GHD did have a consistent impact on health-related quality of life in multivariable models. While adjusting for CRT, age at diagnosis, time since diagnosis, sex and other treatment exposures, GHD was associated with lower reported general health, vitality and mental health, each demonstrating an impact of roughly one-fourth of a standard deviation associated with GHD. Survivors with versus without GHD did not differ on measures of emotional distress, though among survivors without hypothyroidism or hypogonadism, GHD was associated with more symptoms of poor emotional regulation and lower mental health-related quality of life. Thus, the impact of GHD on mental health, combined with an impact on vitality and general health, may suggest a relatively lower perceived well-being or physiologic arousal, rather than physical or emotional limitations.
Hypothyroidism was associated with more self-reported problems in general health and vitality. The effect of hypothyroidism on these outcomes was nearly two-fold larger than the effect of GHD (general health – hypothyroidism impact = −0.51 z-score, GHD impact = −0.26 z-score; vitality – hypothyroidism impact = −0.43 z-score, GHD impact = −0.24 z-score). Hypothyroidism was also associated with worse self-report of bodily pain and social function. Each of these quality of life outcomes demonstrated a non-significant trend for association with GHD. Survivor self-reported task efficiency was also lower in those with hypothyroidism compared to those without. Hypothyroidism following CRT occurred in a subset of those survivors with GHD, though many such survivors with hypothyroidism have GHD.(2) Given the consistently larger effect sizes of hypothyroidism compared to those of GHD, the impact of GHD on self-reported outcomes may be driven more by the combined occurrence of hypothyroidism. Consistent with this conclusion, removal of those survivors with hypothyroidism from multivariable models eliminated the impact of GHD on general health and vitality. The specific impact of hypothyroidism on neurocognitive function and quality of life in the cancer survivor population warrants additional investigation.
In univariate analyses, history of growth hormone replacement therapy, which only occurred in survivors diagnosed with GHD during childhood, was associated with improvement in neurocognitive function. However, these variables impacted self-reported neurocognitive symptoms (problems with memory, organization and emotional regulation), but not direct testing of neurocognitive function. Associations could be due to the apparent effect of GHD on mental health. We have previously reported associations between mental health and self-reported neurocognitive function in long-term survivors of childhood cancer.(20)
Directly tested neurocognitive outcomes were associated with risk factors that have been well-established in the literature. As reported in other studies on long-term survivors of childhood ALL,(1, 21, 22) CRT was consistently associated with all such outcomes. High-dose intravenous methotrexate was associated with lower vocabulary, visual processing speed and fluency, even after adjusting for CRT. High-dose intravenous methotrexate is known to impact white matter integrity and associated processing speed, as well as executive function.(23, 24) Intrathecal methotrexate was not associated with directly tested outcomes; however, since cumulative doses of intrathecal methotrexate are often positively correlated to dose of high dose intravenous methotrexate and negatively correlated to dose of CRT, the effect of intrathecal methotrexate may have been accounted for in these other treatment exposures.
Although this study has many strengths, including objectively assessed neurocognitive outcomes and endocrine status in a large cohort of survivors of a single diagnosis, the findings presented here should be viewed in light of several limitations. First, survivors of ALL are treated with lower doses of CRT compared to survivors of brain tumors. This lower CRT may produce a milder form of GHD. It is possible that more severe GHD following higher doses of CRT may be more clearly associated with neurocognitive function. However, in our experience, adult survivors of CNS tumors treated with higher doses of CRT are at substantially higher risk of neurocognitive impairment, such that a floor effect may limit the ability to detect unique contributions from GHD. Second, GHD was not uniformly assessed during childhood, as it was during adulthood for the participants. Growth hormone testing is typically conducted following observed reduction in stature. Since these survivors were treated with lower CRT doses and thus, may not be at risk for short stature, our classification of childhood versus adulthood diagnosis of GHD likely reflects a distinction between moderate and mild forms of the condition. Third, the diagnosis of GHD in SJLIFE was based on IGF-1 z-scores and not dynamic testing; this may have resulted in the misclassification of participants.(2) Fourth, we may have misclassified some survivors who were not treated with growth hormone replacement. For survivors treated for GHD at our institution, record abstraction would have identified such treatment. However, some survivors may have been treated for GHD at other institutions after graduation from pediatric follow-up care. Such classification would be entirely based on the survivors’ recollection of growth hormone therapy and availability of records from the outside institution. Still, we believe such misclassification to be relatively low.
Limitations notwithstanding, this study demonstrates that survivors of childhood ALL treated with cranial radiation therapy are at higher risk for neurocognitive problems and lower quality of life. GHD may result from cranial irradiation as well, though the neurocognitive problems appear chiefly related to cranial irradiation. Although a randomized clinical trial with neurocognitive testing prior to and following growth hormone supplementation would provide a definitive answer, our results suggest neurocognitive impairment is not related to GHD. These findings have current and future implications for survivors of very high risk and relapsed ALL and survivors of brain tumors, all of whom are treated with cranial irradiation on modern protocols. Long-term emotional adjustment in survivors who develop GHD may be enhanced through early behavioral therapy.
Supplementary Material
Acknowledgments
Supported by the National Cancer Institute (CA195547 to MMH and LLR; CA21765 to Charlie Roberts) and the ALSAC.
Footnotes
Disclosures: The authors have no conflicts of interest to disclose.
References
- 1.Krull KR, Brinkman TM, Li C, Armstrong GT, Ness KK, Srivastava DK, et al. Neurocognitive outcomes decades after treatment for childhood acute lymphoblastic leukemia: a report from the St Jude lifetime cohort study. J Clin Oncol 2013;31(35):4407–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Chemaitilly W, Li Z, Huang S, Ness KK, Clark KL, Green DM, et al. Anterior Hypopituitarism in Adult Survivors of Childhood Cancers Treated With Cranial Radiotherapy: A Report From the St Jude Lifetime Cohort Study. J Clin Oncol 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Waters MJ, Blackmore DG. Growth hormone (GH), brain development and neural stem cells. Pediatric endocrinology reviews : PER 2011;9(2):549–53. [PubMed] [Google Scholar]
- 4.Webb EA, O’Reilly MA, Clayden JD, Seunarine KK, Chong WK, Dale N, et al. Effect of growth hormone deficiency on brain structure, motor function and cognition. Brain : a journal of neurology 2012;135(Pt 1):216–27. [DOI] [PubMed] [Google Scholar]
- 5.Wass JA, Reddy R. Growth hormone and memory. The Journal of endocrinology 2010;207(2):125–6. [DOI] [PubMed] [Google Scholar]
- 6.Lue BH, Huang TS, Chen HJ. Physical distress, emotional status, and quality of life in patients with nasopharyngeal cancer complicated by post-radiotherapy endocrinopathy. Int J Radiat Oncol Biol Phys 2008;70(1):28–34. [DOI] [PubMed] [Google Scholar]
- 7.Huisman J, Aukema EJ, Deijen JB, van Coeverden SC, Kaspers GJ, van der Pal HJ, et al. The usefulness of growth hormone treatment for psychological status in young adult survivors of childhood leukaemia: an open-label study. BMC pediatrics 2008;8:25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hudson MM, Ness KK, Nolan VG, Armstrong GT, Green DM, Morris EB, et al. Prospective medical assessment of adults surviving childhood cancer: study design, cohort characteristics, and feasibility of the St. Jude Lifetime Cohort study. Pediatr Blood Cancer 2011;56(5):825–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hudson MM, Ehrhardt MJ, Bhakta N, Baassiri M, Eissa H, Chemaitilly W, et al. Approach for Classification and Severity Grading of Long-term and Late-Onset Health Events among Childhood Cancer Survivors in the St. Jude Lifetime Cohort. Cancer Epidemiol Biomarkers Prev 2017;26(5):666–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hudson MM, Ness KK, Gurney JG, Mulrooney DA, Chemaitilly W, Krull KR, et al. Clinical ascertainment of health outcomes among adults treated for childhood cancer. JAMA 2013;309(22):2371–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Conners CK. Conners’ Continuous Performance Test II. Noth Tonawanda, NY: Multi-Health Systems, Inc.; 2001. [Google Scholar]
- 12.Wechsler D Wechsler Adult Intelligence Scale - Third Edition. San Antonio, TX: Psychological Corporation; 1997. [Google Scholar]
- 13.Strauss E, Sherman EM, Spreen O. A Compendium of Neuropsychological Tests: Administration, Norms, and Commentary. Third ed. New York: Oxford University Press; 2006. [Google Scholar]
- 14.Delis DC, Kramer JH, Kaplan E, Ober BA. California Verbal Learning Test – Second Edition. San Antonio: The Psychological Corporation; 2000. [Google Scholar]
- 15.Wechsler D Wechsler Abbreviated Scale of Intelligence. San Antonio, TX: Psychological Corporation.; 1999. [Google Scholar]
- 16.Woodcock RW, McGrew KS, Mather N. Woodcock-Johnson III: Tests of Achievement. Itasca, IL: Riverside; 2001. [Google Scholar]
- 17.Krull KR, Gioia G, Ness KK, Ellenberg L, Recklitis C, Leisenring W, et al. Reliability and validity of the Childhood Cancer Survivor Study Neurocognitive Questionnaire. Cancer 2008;113(8):2188–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Derogatis LR. Brief Symptoms Inventory 18: Administration, scoring, and procedures manual. Minneapolis, MN: NCS Pearson Inc.; 2001. [Google Scholar]
- 19.Ware JE, Kosinski M, Gandek B. SF-36 Health Survey: Manual and Interpretation Guide. Lincoln, RI: Quality Metric Incorporated; 2003. [Google Scholar]
- 20.Kadan-Lottick NS, Zeltzer LK, Liu Q, Yasui Y, Ellenberg L, Gioia G, et al. Neurocognitive functioning in adult survivors of childhood non-central nervous system cancers. J Natl Cancer Inst 2010;102(12):881–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Schuitema I, de Sonneville L, Kaspers G, van der Pal H, Uyttebroeck A, van den Bos C, et al. Executive Dysfunction 25 Years after Treatment with Cranial Radiotherapy for Pediatric Lymphoid Malignancies. J Int Neuropsychol Soc 2015;21(9):657–69. [DOI] [PubMed] [Google Scholar]
- 22.Schuitema I, Deprez S, Van Hecke W, Daams M, Uyttebroeck A, Sunaert S, et al. Accelerated aging, decreased white matter integrity, and associated neuropsychological dysfunction 25 years after pediatric lymphoid malignancies. J Clin Oncol 2013;31(27):3378–88. [DOI] [PubMed] [Google Scholar]
- 23.Krull KR, Cheung YT, Liu W, Fellah S, Reddick WE, Brinkman TM, et al. Chemotherapy Pharmacodynamics and Neuroimaging and Neurocognitive Outcomes in Long-Term Survivors of Childhood Acute Lymphoblastic Leukemia. J Clin Oncol 2016;34(22):2644–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Cheung YT, Sabin ND, Reddick WE, Bhojwani D, Liu W, Brinkman TM, et al. Leukoencephalopathy and long-term neurobehavioural, neurocognitive, and brain imaging outcomes in survivors of childhood acute lymphoblastic leukaemia treated with chemotherapy: a longitudinal analysis. Lancet Haematol 2016;3(10):e456–e66. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.