Low Incidence of New-Onset Hypopituitarism After High-Precision Stereotactic Radiation Therapy of Sellar and Perisellar Lesions

Abstract

Purpose

Radiation therapy-induced hypopituitarism (RIH) is a common complication of radiation therapy, even if the pituitary gland is not the primary target but lies within the irradiation field. This study aimed to investigate the impact of high-precision radiation therapy for benign sellar and perisellar tumors, using radiosurgery or fractionated/hypofractionated stereotactic radiation therapy with the Novalis System (Brainlab), on functional pituitary integrity.

Methods and Materials

Fifty-six patients who underwent stereotactic radiation therapy for meningiomas of the sellar or perisellar region of the skull base (n = 36) or pituitary adenomas (n = 20) were recruited between 2013 and 2021 at a single radiooncologic clinic. Fractionated stereotactic radiation therapy was the main irradiation technique (n = 42), followed by hypofractionated stereotactic radiation therapy (n = 4) and stereotactic radiosurgery (n = 10). Irradiation data were assessed retrospectively, and radiation exposure to the pituitary was measured by dose-volume histograms. Each patient received an endocrine diagnostic workup, including a comprehensive functional pituitary analysis.

Results

The median age was 57 years (IQR, 50-63 years), and 42 of 56 patients (75%) were women. RIH was present in 10.7% (n = 6) of patients after a median follow-up of 83.5 months. Gonadotroph function was affected in 7.1% (n = 4) of patients, followed by corticotroph, thyreotroph, somatotroph, and lactotroph function, each at 1.8% (n = 1). Only 1 patient received hormonal replacement for RIH at the study visit.

Conclusions

Awareness of radiation therapy-induced damage to the pituitary gland is essential in the follow-up after radiation therapy of the surrounding structures of the hypothalamic-pituitary axis. High-precision fractionated stereotactic radiation therapy holds a low risk of RIH after irradiation of benign lesions of the sellar and perisellar region.

Introduction

Cranial radiation therapy comprises an effective treatment modality for benign and malignant cranial lesions.1, 2, 3, 4, 5 Radiation-induced damage (RID) to the pituitary gland is a feared consequence of radiation therapy, with incidences of postirradiation hypopituitarism reported between 0% and 89%.^1,3,6, 7, 8, 9, 10, 11, 12 Even if the pituitary is not the primary irradiation target, but located within the radiation field, the mean prevalence of new-onset hypopituitarism was reported to be 66% (range, 55%-76%) after irradiation of nasopharyngeal and intracerebral tumors.⁴ The wide variation of reported radiation therapy-induced hypopituitarism (RIH) might be a result of methodological differences concerning irradiation targets, techniques, and endocrine assessment,^1,4 complicating the risk estimation for the development of RIH.

Skull base tumors of the sellar and perisellar region require special attention because of their anatomic proximity to the pituitary. To date, several factors have been identified that affect the risk of long-term pituitary damage, including radiation dose, fractionation size, and the localization of the irradiation target.^{1,6,8,7,11,13}

The specification of radiation therapy is of critical importance in reducing toxicity to organs at risk. In addition, fractionation of radiation therapy has proven beneficial,⁷ along with recent technical innovations, including stereotactic techniques supported by exact imaging and computerized 3-dimensional (3D) planning, allowing millimetric accuracy through high-conformity application of radiation to a tumor target.¹⁴

Most recently, survival rates of patients with brain tumors have been improved successfully, and patients with benign lesions, such as low-grade meningioma, usually have a normal life expectancy.¹³ Because hypopituitarism may have a negative impact on life expectancy and quality of life,^13,15 a better understanding of RIH is essential, enabling early detection and replacement of possible hormonal deficiencies. For meningioma, the most common nonglial intracranial tumor,² only a few studies with subsequent endocrine follow-up are available, showing a large variation in reported RIH.

Although several studies examined grossly for clinical signs of hypopituitarism and detected none or only rare cases of new RIH,16, 17, 18 dedicated endocrine diagnostic workups were only occasionally performed, for example, by Minniti et al.¹⁹ Using fractionated stereotactic radiation therapy (FSRT), the first manifestation or worsening of pre-existing hypopituitarism occurred in 19% after a median follow-up of 36 months.¹⁹

For fractionated radiation therapy without application of the stereotactic method, but using conventional methods such as 3D and intensity modulated radiation therapy planning, reported RIH varies between 38%⁷ and 60%.⁸

In this mono-centric study, we systematically investigated the impact of high-precision radiation therapy for benign sellar and perisellar tumors using radiosurgery or FSRT/hypofractionated stereotactic radiation therapy (HFSRT) with the Novalis System (Brainlab) on functional pituitary integrity.

The aim of our study was to analyze the risk level of these patients of developing significant functional pituitary defects after stereotactic radiation therapy.

Methods and Materials

Patient selection

The medical database for radiooncology at the Robert Janker Klinik Bonn, a clinic specializing in radiooncological treatment working with an interdisciplinary team of radiooncologists, interventional and diagnostic radiologists/neuroradiologists, medical oncologists, and neurosurgeons, was screened for patients who had received stereotactic radiation therapy with the Novalis System (Brainlab) since 2013. All patients with a benign lesion near the pituitary within the sellar and perisellar region were included and prospectively contacted for comprehensive pituitary testing at the Department of Internal Medicine, Section of Endocrinology, University Hospital Bonn.

Eligibility criteria were as follows: (1) the pituitary gland was within the radiation field, (2) radiation therapy dated back ≥1 year, and (3) aged 18 to 85 years. Exclusion criteria were as follows: (1) anterior pituitary insufficiency of more than 2 axes before irradiation, (2) steroid treatment within the last 12 months, and (3) current pregnancy or lactation.

Blood pressure, pulse rate, body mass index (BMI), and current medication were recorded. Patient histories and reports were screened for previous endocrine records or hormonal laboratory analyses. Clinical status and quality of life (European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Brain Cancer Module 20, EORTC QLQ-BN20, and European Quality of Life 5 Dimensions 5 Level Version, EQ-5D-5L) were evaluated (which will be available in a separate publication).

Patients who were unable to schedule an appointment for diagnostic testing with our department were still included as long as the most recent comprehensive endocrine workup was available. For patients with prior cranial surgery, an endocrine assessment between surgery and radiation therapy was mandatory to establish a preradiation therapy hormonal workup.

The Ethical Committee of the University of Bonn (247/21) and the medical association Nordrhein (2021416) approved this study, which was performed in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained prior to study participation from every patient.

Laboratory assessment

The biochemical assessment of pituitary function included the measurement of the following baseline parameters: thyroid-stimulating hormone (TSH), free thyroxine, luteinizing hormone, follicle-stimulating hormone, estradiol, fasting testosterone and sex hormone-binding globulin in males, prolactin, adrenocorticotropic hormone (ACTH), morning serum cortisol (8-10 AM) (all performed with a chemiluminescence immunoassay, cobas e 801 analyzer, Roche Diagnostics) and 60 minutes after stimulation with 250 µg cosyntropin (Alfasigma S.p.A.), insulin-like growth factor 1 (IGF-1; via immunoassay, IDS iSYS System, IDS PLC), blood count (via fluorescence flow cytometry and photometry, Sysmex XN-9000/XN-1000 analyzer), and serum sodium and potassium (via potentiometry, cobas 8000 ISE analyzer, Roche Diagnostics). Estradiol measurements were taken irrespective of menstrual cycle or menopausal history. The in-house assays were all provided by the central laboratory of the University Hospital Bonn, Germany.

Diagnostic criteria for endocrine assessment

Central hypothyroidism

Central hypothyroidism was diagnosed based on free thyroxine levels below the reference range in conjunction with a low or normal TSH.²⁰

Central adrenal insufficiency

Functional integrity of the hypothalamic-pituitary-adrenal axis was assessed by morning plasma ACTH and serum cortisol levels, in addition to a 250 µg cosyntropin stimulation test under standardized conditions. Central adrenal insufficiency was defined as a low morning cortisol level (<5 µg/dL) in the absence of an elevated ACTH value, or an insufficient rise in cortisol during the cosyntropin stimulation test after 60 minutes (<18.1 µg/dL).^20,21 In case of absent consent or contraindication for the performance of the cosyntropin stimulation test, values of morning cortisol > 15.0 µg/dL were interpreted as sufficient and values <5.0 µg/dL as insufficient for functional adrenal integrity.²⁰

Prolactin deficiency

Lactotroph insufficiency was defined as a prolactin level below the reference range.²⁰ Free prolactin was measured after polyethylene glycol precipitation (PEG) of macroprolactin.

Growth hormone deficiency

Growth hormone deficiency was defined as IGF-1 levels below the age-specific reference range.

Gonadotroph deficiency

Male patients with a morning serum testosterone level or free androgen index below the reference range and a corresponding gonadotropin value in or below the reference range were classified as gonadotropin-insufficient. Postmenopausal women with low estradiol and gonadotropin levels below the postmenopausal reference range were considered gonadotropin-deficient. For premenopausal women, evidence of secondary gonadotropin insufficiency was considered when oligo- or amenorrhea, low estradiol, and low or normal concentrations of follicle-stimulating hormone and luteinizing hormone were present.

Argnine vasopressin (AVP) deficiency

Patients with confirmed polyuria of >50 mL/kg/d were suspected of having central diabetes insipidus and underwent further diagnostic testing.

Radiation therapy

All patients received stereotactic radiation therapy using a 6 MV linear accelerator with a built-in multileaf collimator (Novalis 3m System, BrainLab). The irradiation setup included a thermoplastic stereotactic mask system, a 6-dimensional robotic couch, and daily image guided radiation therapy using ExacTrac positioning (Brainlab). Treatment planning for radiosurgery and FSRT was done with the iPlanRT 3D treatment planning system (Brainlab). First, magnetic resonance imaging/computed tomography fusion planning was performed. The gross tumor volume (GTV) was defined as the region of enhancement on T1-weighted 3D magnetic resonance images (1 mm sliced), in combination with thin-slice (2 mm) coronal T2 images. Organs at risk, including the optic nerves, chiasm, lenses, brainstem, and pituitary, were delineated. The focus was on effective dose distribution enclosing the tumor (ie, GTV). The pituitary was not specifically spared.

In cases of radiosurgery, the prescribed marginal dose ranged from 12 to 18 Gy. No additional margin was added to the GTV for the planning target volume (PTV): ie, PTV = GTV. The PTV was enclosed by 95% to 100% of the prescribed dose. The constraint for the optical system was a maximum dose (Dmax) of 8 Gy, and the mean dose to the pituitary volume was <11 Gy (D50%).

In FSRT, the standard irradiation protocol delivered a total dose of 50.4 Gy in a mean of 28 fractions. A margin of 1 to 2 mm was added to the GTV to define the PTV. The tolerance dose (5/5) for these organs was as follows: 50 Gy for the optic nerves and chiasm, 10 Gy for the lenses, and 50 Gy for the brainstem.

In HFSRT, the standard prescription dose was 40 Gy (IQR, 40-44 Gy) delivered in 10 fractions (IQR, 10-11 fractions). A margin of 1 mm was added to the GTV to define the PTV.

Statistical analysis

Characteristics of study participants are presented as frequencies (percentages) for categorical variables and median (IQR) for continuous variables, as a normal distribution could not be assumed. Group-wise comparisons were performed using Fisher’s exact test or χ² test for categorical variables, and the Mann-Whitney U test for continuous variables. All tests were 2-sided, and a P value of <.05 was considered statistically significant. Statistical analyses were performed using R (version 4.2.1, R Foundation for Statistical Computing).

Results

Among 1517 patients undergoing stereotactic radiation therapy for cranial tumors with the Novalis System (Brainlab) between 2013 and 2021 at the Robert Janker Klinik Bonn, 200 patients were treated for tumors in the sellar and perisellar region. Finally, 64 patients attended the study visit or had recent endocrine results available (n = 7). In 7 of 64 patients, the cause of hypopituitarism (radiation therapy vs cranial surgery) remained unclear because of a lack of preintervention endocrine reports. In addition, 1 patient had proven glucocorticoid-induced adrenal insufficiency. Therefore, these 8 patients were excluded from the final analysis (Fig. 1). As a result, 56 patients were suitable for reliable analysis. The patient characteristics are shown in Table 1.

Figure 1.

Flow chart of patient selection.

Table 1.

Baseline clinical characteristics and patients’ demographics

	Median (IQR) or n (%)
Total no. of patients	56 (100)
Gender
Male	14 (25)
Female	42 (75)
Age at endocrine assessment (y)	57 (50-63)
Follow-up after RT (mo)	83.5 (42-103)
BMI > 30 kg/m2	24 (42.9)
Type of tumor
Meningioma	36 (64.3)
Pituitary adenoma	20 (37.7)
Surgical resection before RT	32 (57.1)
Meningioma, n	12
Pituitary adenoma, n	20
Localization of meningiomas
Sphenoidal	13 (36.1)
Cavernous sinus	8 (22.2)
Clivus/petroclival	4 (1.1)
Frontobasal	3 (8.3)
Tuberculum sellae	2 (5.6)
Optic nerve sheath	2 (5.6)
Petrous ridge	2 (5.6)
Falx	1 (2.8)
Temporal	1 (2.8)
RT modalities
Fractionated stereotactic RT	42 (75)
Radiosurgery	10 (17.9)
Hypofractionated stereotactic RT	4 (7.1)

Abbreviations: BMI = body mass index; RT = radiation therapy.

Baseline patient characteristics

The median (IQR) age was 57 years (50-63 years), and 42 of 56 patients (75%) were woman; 36 of 56 patients (64.3%) had meningiomas, and 37.3% (n = 20) had pituitary adenomas (endocrine inactive n = 13; somatotroph adenomas n = 5; and prolactinoma n = 2).

The majority of patients underwent FSRT (75%, n = 42); the remaining cases received radiosurgery (17.9%, n = 10) or HFSRT (7.1%, n = 4). Medical treatment for endocrine-active pituitary adenomas was present in 2 patients with acromegaly, with somatostatin analogs, and 1 patient with prolactinoma received a dopamine agonist. No patient had received chemotherapy prior to treatment. At the time of the study visit, all patients demonstrated stable tumor control as confirmed during their routine medical follow-up. Baseline and radiological characteristics dependant on irradiation technniques are displayed in Table 2.

Table 2.

Baseline and radiological characteristics dependent on irradiation techniques

Baseline / radiotherapeutical characteristics	Irradiation technique
	FSRT	HFSRT	RS
N	42	4	10
Age (y)	57.0 (50.5-62.0)	57.5 (51.5-61.0)	53.0 (49.0-65.8)
Sex (women)	31 (73.8)	3 (75.0)	8 (80.0)
BMI (kg/m2)	30.3 (24.1-33.3)	33.6 (30.2-35.0)	23.3 (21.7-25.3)
Meningioma	29 (69.0)	2 (50.0)	5 (50.0)
Previous surgery	22 (52.4)	2 (50.0)	8 (80.0)
Previous RT	4 (9.5)	0 (0.0)	2 (20.0)
Follow-up (y)	90 (52-105)	61 (36-90)	60 (38-80)
Total dose	50.4 (50.4-50.4)	42.0 (40.0-44.0)	12.3 (12.0-13.0)
Dose to the pituitary gland
Dose_min	13.2 (7.5-28.4)	6.3 (4.2-8.5)	1.3 (0.7-2.5)
Dose_mean	40.1 (22.2-50.4)	21.3 (14.3-30.4)	7.7 (6.4-8.0)
Dose_max	53.2 (44.4-55.5)	37.3 (31.4-42.2)	13.0 (12.1-14.3)

Data are presented as median (IQR) or n (%).

Abbreviations: BMI = body mass index; FSRT = fractionated stereotactic radiation therapy; HFSRT = hypofractionated stereotactic radiation therapy; RS = radiosurgery; RT = radiation therapy.

Radiation therapy-induced hypopituitarism

Pre-existing hypopituitarism was present in 10 of 56 patients (17.9%) (thyreotroph n = 6; gonadotroph n = 4; and corticotroph n =3). An insufficiency of 2 axes was observed in 3 of 56 patients (5.4%). All of these 10 patients had received cranial surgery prior to radiation therapy, and 9 of them suffered from a pituitary adenoma. No patient showed clinical signs of AVP deficiency syndrome.

At the time of endocrine evaluation, 23 of 56 patients (41.1%) received hormonal replacement therapy: 17 (30.4%) for primary hypothyroidism, 6 (10.7%) for secondary hypothyroidism, and 3 (5.4%) for secondary adrenal insufficiency. Hypogonadism was treated in 2 of 56 patients (3.6%), 1 woman with menopausal symptoms and 1 man with secondary hypogonadism.

In 19 of 56 patients, the cosyntropin stimulation test was not performed because of previously proven corticotroph deficiency (n = 3), missing consent, or multiple allergies (n = 16). Of these, 9 patients had a morning serum cortisol above the predefined cutoff of 15 μg/dL, and the remaining 7 patients presented a morning serum cortisol > 5 μg/dL without clinical signs of hypocortisolism. Thus, no clinically relevant hypocortisolism was assumed.

Evidence of new-onset pituitary deficiency after radiation therapy was found in 6 of 56 patients (10.7%). These 6 patients received FSRT for meningioma (n = 4) or pituitary adenoma (n = 2). Three patients had cranial surgery prior to radiation therapy; none of them had previous irradiation. Gonadotroph function was most frequently affected in 4 of 56 patients (7.1%), followed by corticotroph, thyreotroph, somatotroph, and lactotroph function in 1 patient (1.8%) each (Fig. 2). One female patient presented RIH of 3 axes (corticotroph, gonadotroph, and thyreotroph) after a follow-up of 43 months (mean dose to the pituitary, 47.4 Gy) and previous operative treatment of a pituitary adenoma 4 years before radiation therapy. In the other 5 patients, only 1 hormone axis was affected. Only 1 of 56 patients (1.8%) received hormonal replacement for RIH at the study visit (levothyroxine and hydrocortisone). For 1 patient with a growth hormone (GH)-secreting pituitary adenoma, the diagnosis of RIH was present before the study visit. For the remaining 5 patients, RIH was diagnosed at the time of the study visit or external endocrinologic investigation.

Figure 2.

Overall incidence and affected axes of radiation therapy-induced hypopituitarism for the study cohort of 56 patients.

Subgroup analysis was performed for patients treated with FRST (n = 42): RIH was present in 6 patients (14.3%). The mean radiation dose to the pituitary was 40.0 Gy (Table 3).

Table 3.

Fractionated stereotactic radiation therapy: a subgroup analysis of patient and radiotherapeutic characteristics

Patient / Radiotherapeutic characteristiccs	FSRT (n = 42)			FSRT/meningioma (n = 29)
	RT-induced hypopituitarism			RT-induced hypopituitarism
	No	Yes	P value	No	Yes	P value
N	36	6		25	4
Age (y)	58.0 (50.0-62.3)	56.0 (52.8-57.0)	.773	58.0 (53.0-62.0)	56.00 (54.3-60.0)	1.000
Sex (women), %	25	6	.283	19	4	.663
BMI (kg/m2)	31.2 (25.9-33.6)	23.5 (22.6-28.2)	.131	27.6 (23.7-32.6)	23.1 (21.8-29.6)	.343
Meningioma	25 (69.4%)	4 (66.7%)	1	25 (100%)	4 (100%)	NA
Previous surgery	19 (52.8%)	3 (50%)	1	8 (32%)	1 (25%)	1.000
Previous RT	4 (11.1%)	0 (0%)	.915	3 (12%)	0 (0%)	1.000
Follow-up (y)	84 (54-104)	101 (56-110)	.495	89 (54-101)	103 (81-113)	.312
Dose (Gy)	50.4 (50.4-50.4)	50.4 (50.4-54.0)	.781	50.4 (50.4-50.4)	50.4 (50.4-51.3)	.471
Dose to the pituitary gland (Gy)
Dose_min (Gy, median IQR)	11.0 (5.1-25.9)	28.5 (19.6-34.6)	.091	9.0 (4.3-21.3)	21.1 (16.1-27.3)	.114
Dose_mean (Gy, median, IQR)	37.9 (21.1-49.1)	49.5 (38.9-51.8)	.131	31.3 (16.4-46.9)	43.9 (33.9-52.52)	.164
Dose_max (Gy, median, IQR)	52.9 (42.1-55.3)	54.2 (51.7-55.7)	.297	50.7 (38.7-54.4)	54.5 (52.0-56.4)	.184
RT-induced hypopituitarism
Total, 1 or more axes deficient	-	6 (14.3)		-	4 (13.8)
Gonadotroph	-	4 (9.5)		-	2 (6.9)
Lactotroph	-	1 (2.4)		-	1 (3.4)
Somatotrop	-	1 (2.4)		-	1 (3.4)
Corticotroph	-	1 (2.4)		-	-
Thyreotroph	-	1 (2.4)		-	-

Data are presented as median (IQR) or n (%).

Abbreviations: BMI = body mass index; FSRT = fractionated stereotactic radiation therapy; NA = not available; RT = radiation therapy.

Among patients with meningiomas treated with FSRT (n = 29), the prevalence of new-onset RIH was 13.8% (n = 4).

In both subgroups, patients with new-onset RIH showed a nonsignificant tendency toward a higher mean dose to the pituitary (all FSRT: 37.9 vs 49.5 Gy, P = .131; meningioma and FSRT: 31.3 vs 43.9 Gy, P = .164).

Additionally, there were no significant differences between patients with and without RIH concerning further clinical characteristics, previous surgery or radiation therapy, duration of follow-up, age, sex, BMI, or irradiation target.

After excluding axes that were diagnosed as insufficient before radiation therapy from the data analysis and those under hormonal replacement therapy at the study visit, the rate of new-onset hormone deficiencies changed slightly as follows: whole cohort: gonadotroph 8% (4/50), corticotroph 1.9% (1/52), and thyreotroph 3.1% (1/32). For the subgroup of patients treated with FSRT: gonadotroph 10.5% (4/38), corticotroph 2.6% (1/38), and thyreotroph 4.6% (1/24). For the subgroup of patients with meningiomas treated with FSRT: gonadotroph 8.7% (2/28), corticotroph 0% (0/27), and thyreotroph 0% (0/19).

In patients receiving radiosurgery or HFSRT, no case of RIH was diagnosed. For radiosurgery (n = 4), the median dose to the pituitary was 7.7 Gy (IQR, 6.4, 8.0 Gy); 80% of patients had prior surgery, 50% had pituitary adenomas, and 50% had meningiomas.

Only 4 of 56 patients received HFSRT, with a median dose to the pituitary of 21.3 Gy (IQR, 14.3-30.4 Gy). Half of this group had previous surgery, and 50% were treated for pituitary adenomas and meningioma, respectively. As both groups were relatively small, further statistical analysis was not performed.

Discussion

This study presents a detailed examination of pituitary function in 56 patients who underwent high-precision radiation therapy with a dedicated stereotactic system (Novalis System, Brainlab) at our institution for sellar or perisellar lesions, including meningiomas of the skull base and postoperative residual or recurrent pituitary adenomas. We performed a comprehensive, standardized, prospective endocrine assessment of the pituitary function with a median follow-up of 83.5 months.

Our final analysis included the interpretation of laboratory workup and detailed clinical follow-up. The median age was 57 years, and the study population was predominantly female (75%), which is in accordance with other publications.^7,8

Interestingly, RIH occurred solely in patients who received FSRT and comprised 10.7% (n = 6) of patients and 14.3% in the FSRT subgroup. Only 1 patient in the study cohort received hormone replacement because of RIH.

To the best of our knowledge, endocrine assessment after FSRT for meningiomas in the sellar and perisellar region has not been systematically evaluated before. We found that new RIH in this subgroup was relatively low at 13.8% (n = 4), and no patient received hormone replacement at the time of the study visit for RIH in this subgroup. Therefore, the risk for a clinically relevant RIH was quite low in our cohort.

There is only one study by Minniti et al¹⁹ that analyzed RIH after FSRT for skull base meningiomas, predominantly those of the cavernous sinus. Reported rates of new-onset or worsening of pre-existing hypopituitarism were 19% after a relatively short median follow-up of 36 months. In accordance with the well-known pituitary vulnerability to radiation exposure, the somatotroph axis was mainly affected (17%), followed by the gonadotroph axis (13%) and the thyreotroph and corticotroph axes (8% each).¹⁹ The reported study¹⁹ included only meningiomas of >4 cm in diameter, which represents an independent risk factor for developing RIH.⁸ Therefore, comparison with our study protocol and outcomes is rather limited.

Two recently published studies that provided subsequent endocrine follow-up to detect RIH after nonstereotactic FRT of meningiomas in the sellar and perisellar region reported higher hypopituitarism rates of 38%⁷ to 60%⁸ after a comparable follow-up period. The radiation protocol of Raymond et al⁸ included an additional margin of 1 cm to the GTV (ie, the clinical target volume) plus a PTV of 3 mm. Additionally, one-third of the patients were treated with a conventional 3D planning technique on a linear accelerator instead of intensity modulated radiation therapy, as in the remaining cohort.⁸ These technical differences potentially resulted in a more intense radiation exposure to the pituitary or other ventral structures of the hypothalamic-pituitary axis. Partoune et al⁷ followed a more dedicated algorithm for the diagnosis of TSH deficiency as well as dynamic testing to reveal somatotroph deficiency. This might, in part, explain the higher rates of RIH found in that study⁷ than the present one. Both studies excluded patients with preradiation therapy hypopituitarism.^7,8

Interestingly, these 2 studies reported higher mean doses to the pituitary of 47 Gy⁸ and 48.9 Gy⁷ compared with our study (FSRT, 33.7 Gy). These differences might contribute to the discrepant results, as dose dependency of RIH has been reported previously,^1,6,7 and critical thresholds for the mean radiation dose to the pituitary of ≤45 Gy^7,22 or <50 Gy²³ have been discussed in the literature.

Interestingly, no RIH occurred after radiosurgery and HFSRT. For radiosurgery, this might be explained by low mean radiation doses of 7.7 Gy to the pituitary. In comparison, a recent review reported that hypopituitarism is a common phenomenon, with a pooled estimate of 21% for hypopituitarism after radiosurgery and 3% after HFSRT for nonfunctioning pituitary adenomas.²⁴ After radiosurgery of pituitary adenomas, a mean dose of 11 Gy to the pituitary was reported to represent an important cutoff for the development of RIH.²⁵ The rate of hypopituitarism after 5 years was 5% (<11 Gy) versus 51% (>11 Gy). As noted, only 1 patient in our cohort received more than 11 Gy.

Other reported associations between radiosurgery of pituitary adenomas and new-onset hypopituitarism include the distance between the pituitary and the center of the tumor,²⁶ a Dmax to the pituitary stalk >10 Gy,²⁷ a higher marginal dose to the normal pituitary gland,²⁸ and duration of follow-up.²⁶

An apparent difference from other cohorts is the low rate of growth hormone deficiencies in our study. GH-producing cells are typically the most vulnerable to radiation exposure, followed by the gonadotroph, corticotroph, and thyreotroph axes.¹³ In our study, growth hormone deficiency was assumed when IGF-1 levels were below the age-specific reference range in the context of pituitary radiation exposure. IGF-1 is commonly used as a screening parameter for growth hormone deficiency, but it is prone to diagnostic error, as it is (among other things) dependent on age, liver disease, BMI, and nutritional status. Diagnostic accuracy is markedly heightened by the use of dynamic testing, similar to the insulin tolerance test.²⁹ When dynamic testing was included in the endocrine evaluation after FRT for meningiomas of the sellar and perisellar region, rates of GH deficiencies of 17% to 35% have been described.^6,7,19 Thus, our data possibly underestimates the prevalence of growth hormone deficiency.

In summary, risk stratification of postirradiation hypopituitarism remains extremely challenging. Earlier studies identified several risk factors for the development of hypopituitarism, such as duration of follow-up,^1,7,8,9,12 dose to the pituitary,^6,7 PTV, dose to the hypothalamus, meningioma size, location of meningiomas, and regression of meningiomas after irradiation.⁷

Statistically, we could not demonstrate a significant correlation between RIH and dose to the pituitary, BMI, duration of follow-up, age, sex, previous radiation therapy, previous surgery, or tumor entity across all subgroups. Although there was a clear tendency toward higher mean FSRT doses and developing RID in all patients receiving FSRT (49.5 vs 37.9 Gy), and in the subgroup with meningioma receiving FSRT (43.9 vs 31.3 Gy), statistical significance might have been lost because of the limited sample size.

Only a small fraction of patients developed RID in our cohort; of these, only 1 of 56 patients (1.8%) needed additional hormonal replacement therapy. This data can help inform clinicians and patients in their decision-making, especially young patients with benign disease. Our data also support the use of stereotactic radiation technique with dedicated tools whenever available and appropriate, dose limitation to the pituitary, and to spare whenever feasible.

This study gives insight to the risk of developing RIH after stereotactic radiation therapy in clinical practice and daily life setting, as we included patients with already existing hypopituitarism of up to 2 hormone axes before radiation therapy took place. Our results can help to guide endocrine follow-up and assessment after FSRT.

Limitations

This study has important limitations. First, it is a retrospective study subject to selection bias, and the results should be interpreted as correlations instead of causal relationships. Second, 77 patients who met the inclusion criteria were not enrolled because they were unreachable or declined participation. We included patients with pre-existing pituitary deficiency for statistical analysis; thus, the number of potential new-onset pituitary insufficiencies after radiation therapy was quite limited. Third, 3 radiotherapeutic techniques (FSRT, HFSRT, and RS) were applied to distinct patient populations with differing baseline characteristics. As a result, direct comparisons of outcomes between these modalities are inherently limited, and conclusions regarding the relative safety of HFSRT compared with FSRT or RS should be interpreted with caution. Fourth, not every patient underwent a comprehensive endocrine assessment before radiation therapy, which led to the exclusion of 7 patients from statistical analyses, because the cause of hypopituitarism could not be identified. This exclusion might represent a bias in favor of a lower prevalence of postradiation therapy hypopituitarism in our cohort. Lastly, the functional integrity of the hypothalamus is vulnerable to radiation exposure, and its dose dependency is well known.^7,13,23 The impact of the radiation dose on the hypothalamic area remains unclear in our study. Thus, low events of RIH in our study may reflect a successful protection of the hypothalamus, as a correlation between radiation exposure of the pituitary gland and the hypothalamic area has been demonstrated after FSRT of skull base meningiomas.⁷

Disclosures

None.