Never Forget Germ Cell Tumour in Cranial Diabetes Insipidus: Its Atypical Presentation and Management
Germ cell tumours are frequently associated with endocrine complications, but they may not be evident on the initial cranial imagings. We report an adolescent boy who presented with cranial diabetes insipidus. It was until his sixth set of magnetic resonant imaging which showed the presence of tumour, throughout the period he already developed full-blown panhypopituitarism. Physicians need to be aware of the possibility of germ cell tumours for cranial diabetes insipidus even their first few sets of cranial imagings are not informative. Appropriate hormonal and tumour management are also discussed.
Keyword : Central diabetes insipidus; Germinoma; Panhypopituitarism
A 14-year-old boy (MW) was referred to the paediatric out-patient clinic of a regional hospital in May 2014 for obesity. His weight was 74 kg with body mass index 28.5 kg/m2 and he led a sedentary lifestyle. Upon diet enquiry, his parents incidentally mentioned him to have excessive fluid drinking up to 5 litres per day, and frequent urination with nocturia 1-2 times per night. He had no other urinary symptoms, headache, dizziness or visual concerns. His family history was unremarkable. Physical examination showed an obese boy with good general conditions, full puberty and unremarkable systemic examination.
His initial investigations (Table 1) revealed elevated alanine transferase (ALT), urate, high normal plasma osmolarity and low paired urine osmolarity. These laboratory works were not bothered until 9 months later when MW reported gradual worsening of his polyuria and nocturia. Daily charting of his input and output were up to 4700 and 5500 ml respectively. He felt annoyed if he could not drink adequate volume of fluid. He was instructed to weigh himself at midnight then again with the same scale the next day morning, throughout the night he should refrain from water drinking unless it was irresistible. He was asked to come back for his early morning blood taking paired with urine sampling in the hospital.
This battery of investigations (Table 1) unveiled the biochemical picture of diabetes insipidus (DI) - high plasma sodium (147 mmol/L) and osmolarity (304 mmol/kg) with low urine osmolarity (100 mmol/Kg). He lost 5 lbs (3.2% from his baseline 155 lbs) over the 9-hour water-refraining duration.
He was admitted to ward for further workup of DI. A formal water deprivation test was exempted as the available laboratory results already suggested DI. At that time, all his other hormonal axes remained intact. He showed positive response to sublingual desmopressin (Table 1). Therefore isolated cranial DI was diagnosed.
MW was then evaluated for the possible causes of cranial DI. Most of his subsequent investigations were unrevealing, including tumour markers in blood, skeletal survey, chest X-ray and bone marrow examination. His tumour markers in his cerebrospinal fluid (CSF), however, showed marginally elevated human chorionic gonadotropin (hCG) (1.7 IU/L). His first magnetic resonance imaging (MRI) of his brain and pituitary in March 2015 disclosed absent bright spot in his posterior pituitary in T1 scan, which was a common finding in cranial DI, and a mildly homogenously thickened pituitary stalk (Figure 1).
In view of his cranial DI, mild elevation of CSF hCG and the MRI findings, MW and his parents were interviewed by the endocrinologists, oncologists and neurosurgeons. He was advised to have pituitary biopsy to exclude sinister causes of his cranial DI. Sadly they turned down the suggestions, or even an interim lumbar puncture to assess the progression of tumour markers, as they did not accept the risks of the invasive procedures in spite of full explanation. They only agreed to repeat his MRI as interim surveillance. Thereafter he had done 5 sets of MRI in more than 2 years' time. Although slow progression of the pituitary stalk thickening was evident with the last MRI showing the lesion abutting on the optic chiasma, no new symptoms were reported and no abnormal neurological signs were detected. He tolerated his nocturia with doubtful compliance to desmopressin. Despite detailed explanation of the MRI findings of enlarging lesion and possible disease progression, they insisted on no further investigations but only continual follow up by their private neurosurgeon.
In February 2017, MW agreed to have another hormonal assessment after his frequent defaults of the scheduled blood taking. This morning blood sampling made the worry of the endocrinologists come true. As in Table 1, he eventually developed panhypopituitarism - central hypothyroidism, undetectable morning cortisol and hypogonadotropic hypogonadism. After a lot of negotiation and explanation, MW and his parents finally agreed to admit in March 2017 for hormonal replacement.
He was first replaced with hydrocortisone, with urine output carefully monitored. Before adding thyroxine and testosterone, MW insisted to have earlier discharge after 4 days of stay. He was lost to follow up afterwards. Retrospectively we found he had the next MRI done in the private hospital, further supporting the diagnosis of a pituitary tumour. Transsphenoidal removal of the pituitary tumour was done by their private neurosurgeon in April 2017. Pathological examination reported it to be a germinoma. Thereafter MW was treated with chemo-radiotherapy in another regional hospital, with a full battery of hormonal replacement therapy since then (hydrocortisone, thyroxine, desmopressin and testosterone).
It is not uncommon to see a long-term latency to reach the diagnosis of germ cell tumour (GCT) after cranial DI owing to a multitude of factors, namely lack of significant tumour markers elevation, inconclusive findings in initial MRIs and the indolent symptoms, all of which instigated MW's family to refute more definitive diagnostic strategies like pituitary biopsy and follow-up lumbar punctures even the initial diagnosis of cranial DI was established early.
Central nervous system (CNS) GCTs represent approximately 3% of primary paediatric brain tumours. They commonly originate in the pineal or suprasellar regions, with a few arise in the basal ganglia or thalami.1 The World Health Organization has classified CNSGCTs into two main groups: germinomas and non-germinomatous GCTs (NGGCTs). Two thirds of them are germinomas.
Suprasellar GCTs often present with hormonal abnormalities and visual disturbances. Hypothalamic-pituitary axis disturbance leads to symptoms such as DI, growth failure, delayed or precocious puberty. Endocrinopathies in particular DI often develop ahead of neurological deficits or abnormal imaging. Although MRI is the most reliable tool for radiological detection of GCTs, there is often a time lag, with a median duration of 6 months,2 between the appearance of symptoms and radiological detection.3,4 Delay in diagnosis of GCTs was negatively associated to their overall survival rate,5 and there was a higher incidence of disseminated disease and unfavourable long-term sequelae compared to those with a more timely diagnosis.2
Going back to MW, pituitary stalk thickening could be a non-specific finding, nonetheless, with co-existing cranial DI, more than one third of cases could be due to infiltrative diseases, including GCT, langerhans cell histiocytosis, tuberculosis or hypophysitis.6 A more invasive procedure for a definitive histological diagnosis was justified especially with mildly raised CSF hCG. Eventually it took him more than 2 years to develop full-blown panhypopituitarism after his cranial DI. Throughout the course he only complained of nocturia and fatigue without CNS symptoms. His gradual development of adrenal insufficiency masked his DI symptoms by impairing his free renal water clearance.7 MW's non-specific symptoms together with his partially masked DI made the diagnosis more challenging. On the other hand, it is noteworthy that his TSH was slightly elevated upon presentation of central hypothyroidism. In acquired forms of central hypothyroidism, their TSH concentration could be normal or even slightly elevated, as a result of the qualitative defect in TSH-producing cells secreting TSH isoforms that conserve immunoreactivity but manifest a severe deficiency in intrinsic bioactivity and ability to stimulate TSH receptors.8
In face of treating multiple pituitary hormone deficiencies, it is crucial to replace cortisol ahead of thyroxine, as thyroid hormone accelerates endogenous cortisol clearance and could unmask insufficient cortisol production and precipitate adrenal crisis.9 It is also important to monitor for the development or exacerbation of DI after starting glucocorticoid replacement.
Germinomas are exquisitely radiosensitive and chemosensitive, with cure achieved with radiation alone in 80-90% of patients. Although NGGCTs are less radiosensitive than germinomas and have much worse prognosis, the use of adjuvant chemotherapy has also improved their survival rates, ranging between 40-70%.1 As such, radio-chemotherapy has become the standard treatment for CNSGCTs.10 With clearly characteristic marker elevations in blood and/or CSF and typical radiological findings (like pituitary stalk thickening in our patient), histological examination could be waived to conclude the diagnosis of CNSGCTs. The primary goal of initial surgery should be reserved for biopsy to confirm the diagnosis when pituitary abnormality is evident but normal CSF HCG, as radical resection of intracranial germinomas offers no benefit over chemo-radiotherapy alone.10 Nonetheless, second-look surgery is still recommended if there is residual radiographic lesion after the completion of chemo-radiotherapy, especially in those with persistent positive markers. Hence, our patient might have been exempted from his upfront total tumour resection if the diagnosis of GCT could have been confirmed earlier by a more conclusive elevation of CSF hCG on repeating his lumbar puncture, which could also serve the purpose of waiving the confirmative biopsy of the lesion. Both biopsy and tumour excision are not without significant risk of morbidity. It was however fortunate that during the long time lag, the lesion had not enlarged to an extent to cause permanent vision impairment by compressing on or infiltrating into the optic chiasma.
In summary, this case report highlighted the awareness of GCT as a differential diagnosis of cranial DI. Their symptoms are not always discernible. Serial MRI scans are mandatory to clarify its cause even the initial few scans are not definitively diagnostic. An earlier diagnosis is advisable to avoid CNS damage, permanent visual impairment, pituitary defects and dissemination of tumours. It is important to know the correct sequence of hormonal replacement in hypopituitarism. Finally, for most CNSGCTs, upfront chemo-radiotherapy can usually achieve favourable outcomes.
Declaration of Interest
1. Packer RJ, Cohen BH, Cooney K, Coney K. Intracranial germ cell tumors. The Oncologist 2000;5:312-20.
2. Sethi RV, Marino R, Niemierko A, Tarbell NJ, Yock TI, MacDonald SM. Delayed diagnosis in children with intracranial germ cell tumors. J Pediatr 2013;163:1448-53.
3. Maghnie M, Cosi G, Genovese E, et al. Central diabetes insipidus in children and young adults. N Engl J Med 2000 5;343:998-1007.
4. Mootha SL, Barkovich AJ, Grumbach MM, et al. Idiopathic hypothalamic diabetes insipidus, pituitary stalk thickening, and the occult intracranial germinoma in children and adolescents. J Clin Endocrinol Metab 1997;82:1362-7.
5. Phi JH, Kim SK, Lee YA, et al. Latency of intracranial germ cell tumors and diagnosis delay. Childs Nerv Syst 2013;29:1871-81.
6. Leger J, Velasquez A, Garel C, Hassan M, Czernichow P. Thickened pituitary stalk on magnetic resonance imaging in children with central diabetes insipidus. J Clin Endocrinol Metab 1999;84:1954-60.
7. Linas SL, Berl T, Robertson GL, Aisenbrey GA, Schrier RW, Anderson RJ. Role of vasopressin in the impaired water excretion of glucocorticoid deficiency. Kidney Int 1980;18:58-67
8. Kempers MJE, Lanting CI, van Heijst AFJ, et al. Neonatal screening for congenital hypothyroidism based on thyroxine, thyrotropin, and thyroxine-binding globulin measurement: potentials and pitfalls. J Clin Endocrinol Metab 2006;91:3370-6.
9. Persani L. Clinical review: Central hypothyroidism: pathogenic, diagnostic, and therapeutic challenges. J Clin Endocrinol Metab 2012;97:3068-78.
10. Sawamura Y, de Tribolet N, Ishii N, Abe H. Management of primary intracranial germinomas: diagnostic surgery or radical resection? J Neurosurg 1997;87:262-6.