Abstract
Background/Aim: Hypothalamic-pituitary (HT-P) dysfunction is one of the most common endocrine late effects following cranial radiotherapy. However, there are currently no specific data describing this complication in adult-onset cancer patients after whole brain radiotherapy (WBRT). The present cohort study aims to establish the prevalence of HT-P axis dysfunction in this group of patients. Patients and Methods: Twenty-six cancer patients previously treated with WBRT (median follow-up=20.5 months) received standardized endocrine check-up focusing on HT-P function. Results: In 50% of the patients, impaired hypothalamic-pituitary function was detected during follow-up. While functional loss of a single hormonal axis was evident in 34.6% of patients, 7.7% showed an impairment of multiple endocrine axes, and one patient developed adrenocorticotropic hormone deficiency. Hypothalamic-pituitary dysfunction did not directly correlate with the applied WBRT total doses. Conclusion: In our cohort, hypothalamic-pituitary dysfunction appeared to be common after WBRT and was diagnosed as early as 6 months following radiation. This finding highlights the need for routine endocrine follow-up even in patients with limited life expectancy.
During the last decades, cancer survival rates have increased steadily mainly due to advances in early diagnosis and treatment (1). However, many cancer survivors are affected by therapy-related chronic health conditions. Endocrine sequelae are among the most frequent late effects of cancer therapy and affect up to 50 % of long-term childhood cancer survivors (2). Subsequently, specific long-term follow-up recommendations have been published for pediatric patients to ensure early diagnosis and treatment of endocrine sequelae (3). Only a small number of studies addressed adult-onset cancer survivors and for most cancer entities no long-term follow-up guidelines have been established so far (2).
Whole brain radiotherapy (WBRT) is commonly used in patients with multiple brain metastases or in a prophylactic setting (prophylactic cranial irradiation, PCI) in patients with small cell lung cancer (SCLC) (4). Due to the palliative nature of this therapy, endocrine assessment is mostly not performed. However, it has become obvious that endocrine dysfunction can lead to a reduction in quality of life (QoL), in addition to the potential neurocognitive degeneration induced by radiation (5).
Cranial radiotherapy (CR) represents a major risk factor for the development of hypothalamic-pituitary (HT-P) dysfunction. All hormonal axes of the anterior pituitary gland can be affected, resulting in isolated or combined deficiencies of: i) growth hormone (GH), ii) luteinizing hormone/follicle-stimulating hormone (LH/FSH), iii) thyroid-stimulating hormone (TSH), and iv) adrenocorticotropic hormone (ACTH) (2). Moreover, reduced hypothalamic release of the inhibitory neurotransmitter dopamine may occur and can result in hyperprolactinemia, which contributes to gonadal dysfunction (2). In adult-onset cancer survivors, the overall prevalence of HT-P dysfunction following CR ranges from 20 to 93%, depending on the follow-up time since CR and the applied radiation dose to the HT-P area [reviewed by Mehta et al. (6)].
Currently, there are no specific data on HT-P dysfunction following WBRT. However, the extrapolation of data from patients being treated with CR for non-pituitary brain tumors or nasopharynx carcinomas suggests that current WBRT dose levels are within a range of potential harm (6). Especially for long-term survivors with a good performance status prior to treatment or for small cell lung cancer (SCLC) patients without evidence of disease requiring a prophylactic cranial irradiation (PCI), endocrine dysfunction may be of clinical relevance. As endocrine dysfunction can lead to reduction in QoL, even patients facing limited life expectancy might benefit from a respective follow-up.
To our knowledge, the present cohort study is the first that evaluates the prevalence of HT-P axis dysfunction in adult-onset cancer survivors treated with WBRT. All patients received a standardized endocrine check-up, which, in the absence of specific guidelines for adult-onset cancer survivors, followed recommendations of the current Endocrine Society Clinical Practice guideline for “Hypothalamic-Pituitary and Growth Disorders in Survivors of Childhood Cancer” (3).
Patients and Methods
The local ethical committee of the University of Luebeck (19-409-A) approved the study protocol, which was in accordance with the Declaration of Helsinki. Due to the retrospective and non-interventional nature of the study, requirement of individual informed consent was waived.
All patients were treated with WBRT as part of their cancer treatment at the department of radiation oncology at the University hospital of Luebeck between 2007 and 2018. Inclusion criteria for this retrospective analysis were the following: i) a halted cerebral progress of their disease, ii) a follow-up interval of ≥6 months after completion of WBRT, and iii) a current endocrine check-up including parameters of HT-P function. Exclusion criteria were: i) patients with pituitary metastasis or adenoma and ii) patients receiving dopamine antagonists or agonist. Mainly, WBRT treatment schedules fell into two categories involving either a prophylactic scenario for SCLC patients without evidence of intracranial disease or a therapeutic WBRT with higher doses for patients with brain metastases. Table I summarizes patient- and treatment-related parameters.
Endocrine follow-up was performed in accordance with published guidelines for long-term childhood cancer survivors (3) and was routinely offered to every patient with a stable intracerebral disease surviving ≥6 months following the completion of WBRT. The examination included an early-morning measurement of: i) insulin-like growth factor (IGF1), ii) LH, iii) FSH, iv) testosterone and sex hormone binding globulin (SHBG) (men) or 17 beta-estradiol (women), respectively, v) prolactin, vi) TSH, vii) free thyroxine (fT4), viii) cortisol and ix) ACTH. In cases with indeterminate results, a control examination or a stimulation test (in cases with a suspected ACTH insufficiency) was performed (7). ACTH, cortisol, prolactin, LH, FSH, oestradiol, testosterone, SHBG, TSH and free T4 were measured by electro-chemiluminescence immunoassay (ECLIA); IGF-I was analyzed by chemiluminescence immunoassay (CIA) (both cobas platform, Roche diagnostics, Mannheim, Germany). All assays were performed in the routine clinical biochemistry laboratory at the University Medical Center Schleswig-Holstein, Campus Luebeck, where they have been regularly validated. For the identification of endocrine dysfunction, we applied assay-dependent age and sex-adjusted reference values.
GH deficiency was defined by an IGF-1 level below the reference value. LH/FSH deficiency in males was defined by a low serum testosterone in combination with low to (inappropriately) normal LH/FSH levels. In premenopausal females, LH/FSH deficiency was defined by a low oestradiol level, low or (inappropriately) normal LH/FSH levels and symptoms of oligo-/amenorrhoea, while in postmenopausal women diagnosis was based on the absence of elevated LH/FSH concentration (8). TSH deficiency was defined by low to (inappropriately) normal TSH level in combination with a low serum fT4. The diagnosis of hyperprolactinemia was based on prolactin levels above the upper reference limit (2). ACTH deficiency was suspected in patients with a morning cortisol <140 nmol/l in combination with low to normal ACTH level and confirmed with a standard dose (250 μg) corticotropin stimulation test if peak cortisol levels were below 500 nmol/l (9).
Furthermore, clinical examination and taking the patient's history took place on the day of endocrine assessment. This included an evaluation of the performance status (Karnofsky performance score, KPS), fatigue score (points: 1-10) and National Comprehensive Cancer Network (NCCN) distress thermometer, in which patients were asked to circle the number (range: 0-10) that describes how much distress they experienced during the past week, including the day of assessment (10).
Based on the prevalence of endocrine disorders, we compared age, radiation dose, the interval from WBRT to endocrine assessment and the above-mentioned scores per group using the non-parametric Mann-Whitney U-test, with a significance level of 5% in the IBM SPSS Statistics software for Windows, Version 25.0 (IBM Corp., Armonk, NY, USA). Correlations were analyzed by non-parametric Spearman's Rho test.
Results
A total of 26 patients, 16 females (61.5%), and 10 males (38.5%), were included in this retrospective study. The median age was 58 years (range=36-81) at WBRT and 60 years (range=37-90) at the time of endocrinological examination (Table I). Twelve patients were treated with PCI, and 14 patients received a WBRT with a therapeutic dose for brain metastases. The median interval from WBRT to endocrine examination was 20.5 months (range=6-151) (Table I). Patients with HT-P dysfunction showed a trend towards shorter interval times, with a median interval of 15 months (range=6-132 months) (Figure 1).
HT-P function was impaired in 50.0% (13/26) of the patients (Table II). Nine of 26 patients (34.6%) presented with one HT-P axis dysfunction, and 2 of 26 patients with two (7.7%) or three (7.7%) dysfunctions, respectively. GH deficiency was present in 8.0% of the patients (2/25 patients) (Table II). LH/FSH deficiency was reported in 37.5% (9/24 patients). Five of 14 female patients (35.7 %, missing data for 2 patients) and 4 of 10 male patients (40.0%), respectively, were affected (Table II). Ten of 14 female patients (71.4%) were already postmenopausal, one had premature ovarian failure, one hysterectomy, and two did not experience any cycle irregularities. Two female patients were excluded from the analysis of LH/FSH deficiency; one patient on hormone replacement and another on endocrine therapy. Erectile dysfunction (ED) was reported by 3 of 7 men (42.9%, 3 with missing data); 2 of 3 (66.6%) of these patients were also diagnosed with LH/FSH deficiency.
Hyperprolactinemia was present in 7 of 26 patients (26.9%), of which 3 of 16 were females (18.8%) and 4 of 10 were males (40.0%) (Table II). TSH deficiency was only diagnosed in one patient (1/20, 5.0%); 6 patients already received levothyroxine treatment for hypothyroidism prior to WBRT and thus had to be excluded from the analysis. However, in 2 of 20 patients (10.0%), subclinical hypothyroidism was still present despite their hormone substitution. Furthermore, 2 of 20 patients (10.0%) presented with subclinical hyperthyroidism, characterized by low TSH levels but with fT3 and fT4 levels within the normal range. In one patient (1 of 21, 4.8%; 5 patients with ongoing dexamethasone treatment were excluded from the analysis), we were able to diagnose ACTH deficiency with significantly reduced morning cortisol levels (12 nmol/l). Hydrocortisone replacement was initiated immediately and the diagnosis was confirmed by two consecutive corticotropin stimulation tests; one performed directly after the diagnosis and the second one 3 months later. The patient was instructed regarding stress-dose and emergency glucocorticoid administration and obtained an emergency card and an emergency kit containing injectable high-dose glucocorticoid (3).
Patient assessment scores are given in Table III. At the time of endocrine follow-up, median KPS was 80 (range=50-100). Median fatigue score was 6 (range=1-9) and median points on the NCCN Distress thermometer were 6 (range=2-10). We did not find a significant difference in any of the above scores between patients with or without endocrine deficiencies. Moreover, no significant association was detected between the applied total radiation dose, age, fatigue and distress scores, and the follow-up time from WBRT to endocrine assessment.
Discussion
Endocrine complications following CR are among the most frequent late onset effects of cancer treatment and, as a consequence of improved survival rates (11) affect an increasing number of cancer survivors (2). Although risk-adapted long-term follow-up of childhood cancer survivors has been implemented in the corresponding guidelines for several years, specific recommendations for adult-onset cancer survivors are lacking for most cancer entities. Considering that the number of newly diagnosed (mainly adult-onset) cancer patients exceeded 18 million worldwide in 2018 (12), there is still a large and growing group of cancer survivors with the need of risk-adapted follow-up examinations. Studies focusing on endocrine late effects of CR in adult-onset cancer survivors are rare and mostly based on survivors of non-pituitary brain tumors and nasopharyngeal cancer (13, 14). Currently, no specific data on HT-P dysfunction for adult cancer patients following WBRT exist (6).
The prevalence of hypopituitarism after CR for nasopharyngeal cancer and non-pituitary brain tumors is 20.0 to 93.0% but varies considerably between studies, depending on radiation dose to the HT-P area and the follow-up time since CR (6, 13). In this study, we demonstrated that half of the patients treated with WBRT developed at least one HT-P axis dysfunction and approximately 8.0 % were affected by multiple HT-P axis deficiencies initially diagnosed during routine follow-up.
LH/FSH deficiencies and hyperprolactinemia were the most common complications in our cohort, whereas only 8.0% presented with GH deficiency, as defined by low IGF-1. Although the somatotropic axis is considered as the most vulnerable one to radiation damage, GH deficiency in adult-onset cancer survivors seems less common compared to childhood cancer survivors (2). However, in a study by Kyriakakis et al., which assessed pituitary dysfunction following CR in adult-onset non-pituitary brain tumors by performing regular dynamic pituitary testing, partial or severe GH deficiency was present in 86.9% of patients (15). As normal IGF1 levels do not exclude GH deficiency, the percentage of study patients affected by GH deficiency may have been higher than reported by us, as no routine stimulation testing was performed here. It is important to emphasize that, although GH deficiency in adults is associated with an impaired quality of life, GH replacement therapy in adult cancer survivors is controversial in terms of a potentially elevated risk for cancer recurrence and should thus remain a case-by-case decision, depending on the time of relapse-free follow-up and extent of impairment (2, 15).
Testosterone deficiency as a result of hypogonadotropic hypogonadism may present with depression symptoms, decreased muscle mass, increased abdominal fat and decreased energy as well as with typical signs of gonadal dysfunction (16). This may possibly be aggravated by concomitant hyperprolactinemia, which typically affects about one third of patients after CR (13, 17) and 27% of our study participants. ED can be caused by various conditions, but is also a common symptom of testosterone deficiency (18). This was reported in 42.9% of our male participants of which two thirds were also diagnosed with LH/FSH deficiency. This is in line with results from a previous study assessing endocrine late effects in a larger cohort of adults treated for brain tumors, Hodgkin and non-Hodgkin lymphoma, in which ED was reported in 41.1% of male participants (19). In females, premature menopause as a consequence of LH/FSH deficiency is associated with osteoporosis, an increased risk of atherosclerosis and with symptoms of depression (2). Sex hormone substitution is generally recommended in these patients to avoid cardiovascular complications and reduced bone mass (2). Moreover, hypogonadism was already suggested as a possible contributor to increased fatigue and weakness after radiotherapy (16), which was also demonstrated in our study. Consequently, regular assessment of sex hormone levels in patients following WBRT can facilitate early diagnosis and treatment of hypogonadism and contribute to an improved QoL.
Interestingly, we were able to diagnose ACTH deficiency in an asymptomatic patient as part of the endocrine check-up. Clinical signs of adrenal insufficiency, such as weakness/fatigue, susceptibility to infections or abdominal pain/vomiting were absent. The patient was also affected by TSH deficiency and levothyroxine treatment was initiated several weeks after the beginning of hydrocortisone replacement therapy. ACTH deficiency is the most life-threatening HT-P deficiency and an early diagnosis and start of replacement therapy are essential to avoid adrenal crisis (8).
Noteworthy, our study focused on the detection of secondary hypothyreoidism, meanwhile 20.0% of study participants presented with subclinical primary hyper- or hypothyroidism. When adding patients who already received levothyroxine treatment due to previously diagnosed hypothyroidism (23%) plus one patient (4%) with TSH deficiency, almost half of the study population was affected by thyroid dysfunctions, clearly exceeding the prevalence of these disorders in the general population (20).
Our results emphasize the need for a routine endocrine follow-up of patients treated with WBRT within their first post-treatment year. Our cohort showed HT-P axis dysfunctions as early as 6 months after the end of radiation therapy. In contrast to CR studies that demonstrate dose dependency (21-23), no such correlation was found in our cohort following WBRT, further arguing for shorter endocrine follow-up intervals. In line with this finding, the median follow-up time from WBRT to endocrine assessment of our whole study collective was 20.5 months, while patients with a HT-P dysfunction had a trend towards shorter median follow-up times of 15 months.
Depending on the study, a distress score of more than 3-4 indicates patients with clinically elevated levels of distress (24, 25). Mock et al. rated fatigue in a score of 1-10 (26). The distress and fatigue scores of our collective were elevated with a median of 6. The finding that these scores were not significantly different between patients with and without endocrine deficiencies should not discourage from endocrine testing. Quite the contrary, treatment of these deficiencies might improve QoL in affected patients, a prospective effect that was not encompassed by the retrospective design of our study.
Our retrospective cohort study has several limitations. Although all patients treated with WBRT for more than a decade were considered as potential study participants, only 26 patients met the inclusion criteria. In this respect, overrepresentation of hypothyroidism might represent an oversampling bias of our cohort study. Although the radiation dose to the HT-P area has been demonstrated to be a major risk factor for the development of HT-P complications (2), there was no significant difference regarding the radiation dose in patients with or without an HT-P axis dysfunction. Furthermore, diagnosis of HT-P dysfunction was mainly based on single evaluations of hormone levels. In most cases, the HT-P dysfunction has not been validated by stimulation tests. However, this approach has been suggested in a review of Fernandez et al. summarizing evidence for basal hormone measurements in the evaluation of HT-P axis dysfunctions and has been (partially) incorporated into the current HT-P guidelines for childhood cancer survivors (3, 7). Moreover, our pretreatment endocrine assessment was only based on retrospective records of clinically evident HT-P dysfunction or intake of respective medication and did not include the determination of hormone levels before the start of radiotherapy.
To our knowledge, this study is the first to assess HT-P function in cancer survivors treated with WBRT. We could demonstrate that HT-P axis dysfunctions develop frequently in these patients and can occur early following WBRT. Thus, patients with limited life expectancy may also be affected. Therefore, routine endocrine follow-up is essential to ensure early diagnosis and treatment of endocrine complications to maintain QoL. Moreover, sparing of the HT-P axis like sparing of the hippocampus (5) could be of value as no threshold radiation dose to develop HT-P dysfunction is established and therefore the dose to this area should be as low as reasonably achievable. Our group already showed feasibility of such a sparing approach using a VMAT technique (27). Against the background of a low prevalence of metastases within the HT-P region (28), we plan a prospective trial to further investigate this approach and overcome the limitations of this retrospective study.
Footnotes
Authors' Contributions
JG, SS, DR, and FF participated in the design of the study, SJ, PM, and JG provided data. SJ, FF, SS, JG and PM performed the analyses of the data. SJ and DR drafted the manuscript, which has been reviewed and approved by all Authors.
Conflicts of Interest
On behalf of all Authors, the corresponding author states that there are no conflicts of interest related to this study.
- Received July 6, 2020.
- Revision received August 5, 2020.
- Accepted August 6, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved