Radiological Surveillance for Pituitary Adenomas in Multiple Endocrine Neoplasia Type 1: A Longitudinal Cohort Study

Abstract

Context

Patients with multiple endocrine neoplasia type 1 (MEN1) are genetically predisposed to developing tumors of the parathyroid, pancreas, and anterior pituitary. While widespread genetic testing has increased early diagnoses, the value of radiological surveillance for pituitary adenomas in these patients remains unclear.

Objective

To investigate the impact of pituitary magnetic resonance imaging (MRI) findings on patient management in MEN1.

Design

Retrospective longitudinal cohort study.

Setting

NIH Clinical Center.

Patients

Genetically confirmed MEN1 evaluated between 1977 and September 2024.

Measures

Cumulative proportion and rate of intervention for nonfunctional adenomas.

Results

Three hundred twenty patients with a mean follow-up duration of 14 ± 12 years were identified. One hundred eighty-two out of 320 patients (57%) had pituitary adenomas, of whom 95/182 (52%) had nonfunctional adenomas. All patients with adenoma had biochemical evaluation at the time of pituitary MRI. There were 21/95 (22%) patients with nonfunctional adenomas that needed an intervention throughout their follow-up. The rate of intervention per 100 patient-years for nonfunctional adenomas was 3.1% (confidence interval 2.0-4.7%). All patients with nonfunctional adenoma who required an intervention reported symptoms or had an abnormal pituitary function test. Tumor volume of 38/48 (79%) nonfunctional adenomas followed over a mean of 8.3 ± 6.1 years remained stable. Only 42/518 (8.1%) MRIs for nonfunctional adenomas were prompted by clinical or biochemical abnormalities. Individualized imaging could have avoided approximately 4 MRIs per patient with nonfunctional microadenoma over their follow-up.

Conclusion

In an overwhelming majority of patients with MEN1, radiological surveillance for nonfunctional pituitary adenomas offers minimal added benefit beyond patient-reported symptoms and pituitary function testing.

Multiple endocrine neoplasia type 1 (MEN1) is a genetic disorder characterized by tumors in multiple endocrine tissues, including the anterior pituitary, parathyroid, duodenum, and pancreas. Its prevalence is estimated at 3 to 20/100 000 (1). It is caused by a germline heterozygous inactivating variant in the MEN1 gene with a subsequent somatic “second hit,” which results in loss of the encoded tumor suppressor protein Menin in affected cells and their neoplastic progeny (2, 3). The disease is inherited in an autosomal dominant manner, meaning that first-degree relatives of an affected individual have a 50% risk of carrying the genetic mutation. Given the high disease penetrance, genetic testing is recognized as the gold-standard diagnostic criteria for the disease. In addition to the classic endocrine tumors (anterior pituitary, parathyroid, and entero-pancreatic cells), patients with MEN1 are being increasingly recognized to harbor tumors in additional endocrine and nonendocrine tissues such as thymus, lung, stomach (type 2 gastric neuroendocrine tumors), adrenal cortex and medulla, skin (angiofibromas or collagenomas), meninges, ependymal cells of the brain and spinal cord, smooth muscles (leiomyomas), and adipocytes (1, 4). Thus, patients with the disease are currently recommended to undergo an extensive panel of laboratory assessments and imaging for surveillance of disease-related manifestations (5, 6).

Pituitary adenomas have been reported in 15% to 50% of patients with MEN1 diagnosed at ages 5 to 90 years with a mean age at diagnosis of 38 years (5, 6). For reference, autopsy and imaging studies of persons without a known history of pituitary disease show that 10% to 15% of the population harbor undiagnosed pituitary tumors (7). However, these adenomas affect health outcomes in only about 1 in 1000 persons in the community (7). Data regarding the impact of longitudinal surveillance for MEN1-related pituitary adenomas on patient management is not available. Functional adenomas, particularly prolactinomas and nonfunctional microadenomas, constitute 60% to 95% of MEN1-related pituitary tumors (8, 9). While surveillance for functional pituitary tumors can be performed through laboratory assessments, surveillance for a nonfunctional pituitary adenoma requires a magnetic resonance imaging (MRI) of the pituitary, which can be burdensome with respect to expense, time, and access. There is also a risk of iatrogenic kidney injury and allergic reactions from the gadolinium-based contrast agent used to obtain these images (10). In view of the widening accessibility of genetic testing, patients are being increasingly diagnosed with MEN1 before they develop any disease-related manifestations, resulting in a broader scope of disease surveillance (6). This can place a significant physical, psychological, and financial burden on patients and strain healthcare resources. Furthermore, while the diagnosis of a functional pituitary adenoma typically impacts patient management necessitating medical or surgical intervention, most nonfunctional adenomas do not impact patient management and are managed conservatively through observation until these place the patient at significant risk of mass effect resulting in visual disturbances, frequent headaches, or hypopituitarism (11, 12).

The current recommendation for radiological surveillance of sporadic nonfunctional pituitary microadenomas is cessation of surveillance after 1 to 3 years of stability, while lifelong surveillance with an individualized strategy is recommended for macroadenomas (13). Whether a similar approach can be used for MEN1-related nonfunctional adenomas remains unclear. Currently, a pituitary MRI is recommended at initial evaluation and every 3 to 5 years in patients with MEN1 (5, 6). This recommendation stems from suggestions that MEN1-related pituitary tumors have a more aggressive course with a larger size and a reduced response to therapy (8, 9, 14, 15). However, recent studies highlight an indolent course of MEN1-related pituitary adenomas, raising uncertainty regarding the benefits of radiological surveillance in this patient cohort (9, 14, 16, 17). The National Institute of Diabetes and Digestive and Kidney Diseases has a long-standing clinical research protocol investigating MEN1 (NCT04969926). This research led to the identification of germline variants in the MEN1 gene as the cause of the syndrome in 1997 (2, 3). In this study, we investigate the impact of a diagnosis of MEN1-related pituitary adenomas on patient management and characterize their natural history. We also explore the frequency of asymptomatic nonfunctional adenomas that necessitated an intervention. By utilizing the existing longitudinal data on a large comprehensively phenotyped cohort of patients with MEN1, we aim to assess the benefit of radiological surveillance on clinical management.

Methods

Study Design and Oversight

This single-center retrospective cohort study took advantage of existing data accrued over several decades at the world's largest biomedical research institution. We identified patients with a diagnosis of genetically confirmed MEN1 seen under the institutional review board-approved Natural History Study of Parathyroid Disorders protocol (NCT04969926) between 1977 and September 2024 conducted at the National Institutes of Health (NIH) Clinical Center (Bethesda, MD) and subsequently extracted data pertaining to their genetic testing and clinical course of pituitary manifestations using Research Electronic Data Capture. Written informed consent was obtained from all patients included in this analysis.

Patient Cohort

Participants included patients with genetically confirmed MEN1 and available pituitary imaging. After initial presentation, follow-up frequency was individualized based on the overall status of their MEN1-related manifestations. Pituitary MRI was obtained in all patients at their initial presentation and subsequently at an interval based on clinical judgement, patient preference, and standard of care at the time. Demographic, baseline characteristics, symptomatology (specifically presence of frequent headaches and visual disturbances), pituitary function testing, results of pituitary MRI, treatment history, and histopathology in case of surgery were collected. Pituitary function testing and imaging obtained within ±3 months of a patient encounter were considered to represent the same time point. All patients included in the longitudinal analysis had corresponding pituitary function tests for each visit. The composition of the pituitary function tests varied but at the very least included a serum prolactin ± IGF-1 at every visit. Available pituitary function tests were reviewed and categorized as appropriate or inappropriate, as assessed by an endocrinologist. Dynamic testing for a select axis was performed only when initial test results were abnormal. Medical decision-making for each visit was grouped as positive (initiation, continuation, or cessation of medical therapy, referral for surgery or radiation therapy) or negative (continued observation) for intervention. Tumor volume for nonfunctional pituitary adenomas was independently measured by 2 radiologists, with interobserver differences resolved by consensus. Only patients with at least 2 imaging studies spaced at least 1 year apart were included for the tumor volume analysis. Tumor volumes were estimated using 3-dimensional measurements (length, width, and height). Cases lacking complete measurements were excluded from analysis. An average change in tumor volume (slope) of <10% was classified as stable, ≥20% as growth, and 10% to <20% as indeterminate. These categories are based on RECIST (Response Evaluation Criteria in Solid Tumors) guidelines (18, 19), as no standardized tumor volume grading system currently exists for pituitary adenomas.

Outcomes

Primary study outcomes were the cumulative proportion of intervention for nonfunctional adenomas, rate of intervention for nonfunctional pituitary adenoma, frequency of asymptomatic pituitary adenomas impacting patient management, and rate of tumor growth. Secondary outcomes of interest included cumulative proportion of patients with adenomas, cumulative proportion of patients with adenomas who required an intervention, and time to intervention for the adenoma since its diagnosis.

Statistical Analysis

Categorical variables were summarized as counts and percentages. Continuous variables were reported as mean ± SD and median with interquartile range (IQR). P-values were calculated using chi-square, analysis of variance, or t-tests, as appropriate. Kaplan-Meier survival curves based on time from diagnosis to intervention were calculated to estimate the cumulative proportion of patients needing an intervention. Cumulative incidence rates for the competing risks on functional and nonfunctional adenomas were estimated using the method described by Fine and Gray (20). Poisson regression was used to estimate the intervention rate per 100 patient years of follow-up. The univariable Cox proportional hazard model was used to evaluate predictors of tumors requiring any intervention (age, sex, and tumor size). Tumor volume over the course of follow-up was described using spaghetti plots. The slope of tumor volume change over time was calculated using linear regression for each patient and was divided by the initial volume to calculate percent change in volume. Statistical analysis was performed in R studio (cmprsk and survival packages) and SAS 9.4 (Genmod and Phreg procedures).

Results

Patient and Tumor Characteristics

We identified 320 patients with genetically confirmed MEN1: 161 females (50%) and 159 males (50%). Mean age at diagnosis was 33 ± 16 years, and mean age at last follow-up was 47 ± 16 years. Mean duration of follow-up was 14 ± 12 years(median 12; IQR 3-22). Thirty-eight of 320 patients (12%) were index cases of MEN1 in their respective kindreds. Overall, 182/320 patients (57%) were diagnosed with a pituitary adenoma. Of these, the pituitary adenoma was diagnosed at the time of diagnosis of MEN1 in 78/182 (43%) patients. There were 38/182 (21%) patients with multifocal pituitary adenomas, of which 13/38 (34%) were metachronous. Overall, 87/182 (48%) patients had functional adenomas [78/87 (90%) prolactinomas, 2/87 (2%) GH secreting, 8/87 (9%) ACTH secreting, 1/87 (1.2%) thyrotropic hormone secreting, and 1/87 (1.2%) prolactin-GH cosecreting], and the remaining 95/182 (52%) patients had nonfunctional adenomas. In total, there were 72/182 (40%) macroadenomas with 20/72 (28%) being completely intrasellar without compression of the optic chiasm. Only 9/182 (5%) patients with microadenomas at diagnosis progressed to macroadenomas during the follow-up period. These data are summarized in Table 1.

Table 1.

Demographics

Variables	All patients	No adenoma	Nonfunctional adenoma	Functional adenoma	P-values
Number of patients	320	138	95	87
Age last visit (years)a	47 (16)	47 (17)	47 (17)	45 (16)	.47
	46 (33-60)	48 (33-61)	46 (34-60)	44 (31-60)
Years from first to last visita	14 (12)	11 (12)	14 (11)	18 (12)	<.0001
	12 (3-22)	7 (0.5-20)	12 (4-21)	17 (8-28)
Female sexb	161 (50)	68 (49)	45 (47)	48 (55)	.55
Racec
White	278 (87)	125 (91)	79 (83)	74 (85)	.81
Not reported	16 (5.0)	3 (2.2)	3 (3.2)	5 (5.8)
Black/African American	12 (3.8)	4 (2.9)	3 (3.2)	5 (5.8)
Multiracial	7 (2.2)	3 (2.2)	3 (3.2)	1 (1.2)
Asian	5 (1.6)	2 (1.5)	2 (2.1)	1 (1.1)
American Indian/Pacific Islander	2 (0.6)	1 (0.7)	0 (0)	1 (1.2)
Patients with adenomas
Number of patients	182	NA	95	87
Age at diagnosis (years)d	33 (15)		38 (15)	28 (13)	<.0001
	31 (22-24)		36 (28-46)	26 (17-36)
Macro-adenomad	72 (40)		23 (24)	49 (56)	<.0001
Prolactinomae	78 (43)		NA	78 (90)	NA
Interventionsf
Number of patients with intervention	105		21	84
Rate per 100 patient years	11.00 (9.08-13.32)		3.05 (1.99-4.67)	31.69 (25.59-39.24)	<.0001
Cumulative percent 1 year after diagnosis	41 (34-48)		13 (6-20)	71 (62-81)	<.0001
Cumulative percent 5 years after diagnosis	48 (41-56)		19 (10-27)	79 (71-88)	<.0001
Cumulative percent with intervention at age 40	46 (38-53)		15 (7-23)	76 (66-85)	<.0001
25th percentile for age at intervention	28 (25-31)		49 (39-68)	20 (17-25)	<.0001
50th percentile for age at intervention	43 (38-48)		NA	30 (25-31)	NA

Abbreviations: NA, not available.

^a Mean (SD), median (interquartile range); P-value from ANOVA.

^b n (%); P-value from a chi-square test.

^c n (%); P-value from a chi-square test for White vs non-White.

^d Mean (SD), median (interquartile range); P-value t-test.

^e n (%); P-value from a chi-square test; not applicable for prolactinoma because prolactinomas are part of the definition of functional adenomas.

^f Rate, cumulative percent, and percentiles with 95% confidence interval from Poisson regression or Kaplan-Meier survival curves; 50th percentile is not estimable for nonfunctional adenomas.

Cumulative Proportion of Pituitary Adenomas in Patients with MEN1

The cumulative proportions of all pituitary adenomas was 13% [95% confidence interval (CI) 9-16%] by age 20, 27% (22-32%) by age 30, 45% (40-51%) by age 40, and 58% (52-64%) by age 50 (Fig. 1A). The cumulative proportions of nonfunctional adenomas were 3% (1-5%) by age 20, 10% (7-14%) by age 30, and 22% (17-27%) by age 40 (Fig. 1B). Of the 95 nonfunctional adenomas, 23/95 (24%) were macroadenomas at diagnosis.

Figure 1.

Cumulative proportion of pituitary adenomas in MEN1. (A) Diagnosis ranged from ages 5 to 80 years. Cumulative proportion of adenoma was 13% (95% CI 9-16%) by age 20, 27% (CI 22-32%) by age 30, 45% by age 40 (CI 40-51%), and 58% (CI 52-64%) by age 50 years. (B) Cumulative proportion of adenoma stratified by functional status showing functional adenomas present earlier and are diagnosed at a younger age.

Abbreviations: CI, confidence interval; MEN1, multiple endocrine neoplasia type 1.

Cumulative Proportion of Intervention for MEN1-related Pituitary Adenomas

Overall, 105/182 (58%) patients required medical, surgical, or radiotherapeutic intervention for pituitary adenoma. The cumulative proportion of intervention for pituitary adenomas was 12% (95% CI: 7-17%), 28% (21-35%), and 46% (38-53%) by ages 20, 30, and 40 years, respectively (Fig. 2A). The rate of intervention did not differ by age or sex (Table 2). Of the 105 patients who needed an intervention, 84 patients had a medical intervention for their pituitary adenomas [82/84 (98%) with dopamine agonist], 49 (47%) patients required surgery [7/49 (14%) patients required multiple surgeries], and 12/105 (11%) required radiation therapy [4/12 (33%) with nonfunctioning adenomas]. Mean age at surgery was 32 ± 1 5 years. The most frequent indication for surgery was compression of the optic chiasm or the presence of a functioning adenoma, which together accounted for 58/60 (97%) surgeries. Postsurgical complications included panhypopituitarism (8/60, 13%); cerebrospinal fluid leak (4/60, 7%); cranial nerve injury (3/60, 5%); arginine vasopressin deficiency, formerly called central diabetes insipidus (2/60, 3%); and infection (1/60, 2%).

Figure 2.

Cumulative proportion of intervention for pituitary adenomas in MEN1. (A) All pituitary adenomas by age at intervention. (B) Nonfunctional adenomas by age at intervention. (C) Nonfunctional adenomas by years since diagnosis. (D) Nonfunctional adenomas by years since diagnosis, stratified by adenoma size. Lines that do not start at 0 indicate interventions that occurred at or shortly after the time of diagnosis.

Abbreviations: MEN1, multiple endocrine neoplasia type 1.

Table 2.

Rate of intervention for patients with nonfunctional adenomas

		Time from diagnosis to intervention
				Rate per 100 patient years			Life table		Cox regression hazard ratioa
		n patients	n with intervention	Rate	Lower 95% CI	Upper 95% CI	Cumulative percent at 5 years	P-value	Hazard ratio	Lower 95% CI	Upper 95% CI	P-value
Sex	Male	50	9	2.38	1.24	4.57	16.6		Ref
	Female	45	12	3.86	2.19	6.79	21.0	.34	1.51	0.64	3.59	.35
Age	<30	9	2	2.78	0.69	11.10	12.5		Ref
	20-29	23	3	1.63	0.52	5.04	9.4		0.58	0.10	3.50	.55
	30-39	31	10	4.43	2.39	8.24	26.8		1.59	0.35	7.34	.55
	≥40	32	6	2.89	1.30	6.44	18.3	.40	0.95	0.19	4.74	.95
Size	Micro	72	2	0.35	0.09	1.39	3.2		Ref
	Macro	23	19	16.8	10.71	26.33	62.3	<.0001	39.71	9.22	170.98	<.0001

Abbreviations: CI, confidence interval.

^a Univariable Cox regression P-value.

Of the functional adenomas, 75/78 (96%) prolactinomas had an intervention, predominantly using dopamine agonists, of which 22/75 (29%) patients needed surgery and 3/75 (4%) patients required a second surgery; 2 of those ultimately required radiation treatment.

There were 21/95 (22%) patients with nonfunctional adenomas who needed an intervention throughout their follow-up. As expected, patients with functional adenomas were more likely to need an intervention than patients with nonfunctional adenomas (hazard ratio 6.8, 95% CI 4.2-10.9, P < .0001). For patients with nonfunctional adenomas, the cumulative proportion of intervention was 3.4% (95% CI 0.0-7.1%) by age 30 years, 5.9% (95% CI 0.9-11.0%) by age 35, and 14.9% (95% CI 6.7-23.2%) by age 40 (Fig. 2B).

When measured as time from diagnosis, the rate of intervention per 100 patient-years for the 95 patients with nonfunctional adenoma in the cohort was 3.1% (95% CI 2.0-4.7). The cumulative proportion of patients with nonfunctional adenoma who required an intervention within 5 years of its diagnosis was 19% (Table 1, Fig. 2C). When comparing patients according to size of the adenoma, the 23 patients with nonfunctional macroadenomas were more likely to need an intervention than the 72 patients with nonfunctional microadenomas (hazard ratio 39.7, 95% CI 9.2-171.0, P < .0001) (Table 2). The cumulative proportion of patients who required an intervention within 5 years of diagnosis was 62.3% for nonfunctioning macroadenomas and 3.2% for nonfunctioning microadenomas (P < .0001) (Fig. 2D). Among the patients with nonfunctional adenomas requiring intervention, all patients had biochemical abnormalities or symptoms preceding need for an intervention.

There were 2 patients with nonfunctional microadenomas who underwent intervention in our cohort. The first patient, DK-0142, is a 39-year-old female who underwent transsphenoidal surgery for hypercortisolemia, which was later confirmed to be of adrenal origin; immunohistochemistry of resected pituitary was negative for ACTH. The second patient, DK-0422, is a 41-year-old male who underwent transsphenoidal surgery at an outside institution for an unclear indication.

Rate of Tumor Growth

There were 48/95 (51%) patients with nonfunctioning adenomas with 2 or more available tumor volume measurements at least a year apart. The mean duration of follow-up for patients included in this analysis was 8.3 ± 6.1 years. Of these, 6/48 (13%) patients required an intervention for their adenomas, and 3/48 (6%) experienced apoplexy. A spaghetti plot analysis of tumor volumes for nonfunctional tumors without any intervention demonstrated a relatively flat line, indicating minimal annual growth (Fig. 3). Thirty-eight of 48 (79%) adenomas showed stable tumor volumes (±10% change in tumor volume); 3/48 (6%) were indeterminate (10-<20% increase in tumor volume), while 7/48 (15%) showed tumor progression (≥20% increase in tumor volume) slope (Table 3). Due to the small sample size of nonfunctional tumors needing an intervention, we were unable to perform modeling to identify predictors of tumors that showed progression.

Figure 3.

Tumor volumetrics of MEN1-related nonfunctional pituitary adenomas: spaghetti plot of tumor volumes for nonfunctional pituitary adenomas. Red lines indicate patients with pituitary apoplexy; green lines represent those who underwent intervention (volumes measured up to the first intervention); blue lines represent patients with no intervention. (A) Tumor volume over time, by years since diagnosis. (B) Tumor volume over time, by patient age at each visit.

Abbreviations: MEN1, multiple endocrine neoplasia type 1.

Table 3.

Tumor volume slopes for patients with nonfunctional adenomas

	All volumesa	Volumes within 5 yearsb	Change first to second volumec
Number of patients	48	45	38
Age at first volume (years)	38 (14)	37 (14)	38 (13)
	37 (28 to 47)	37 (27 to 45)	37 (29 to 48)
Number of volumes	5.6 (4.2)	3.7 (1.4)	2 (0)
	4 (2.5 to 7)	3 (2 to 5)
Years from first to last	8.3 (6.1)	4.0 (1.2)	1.6 (0.6)
	6.6 (3.5 to 10.6)	4.3 (3.1 to 5.0)	1.4 (1.2 to 2.1)
First volume (mm3)	560 (2057)	560 (2119)	492 (2148)
	64 (33 to 211)	64 (36 to 198)	64 (30 to 180)
Slope per year	37 (240)	2.7 (178)	22 (60)
	0.3 (−0.4 to 8.6) _	0.0 (−0.7 to 3.3)	0 (0 to 2.8)
Slope as % of first volume	17.1 (62.3)	14.4 (63.7)	7.6 (14.1)
	0.3 (−0.1 to 8.3)	0.0 (−0.4 to 8.8)	0 (0 to 6.3)
Slope categoriesd
Improved or worse by <10%	38 (79)	34 (76)	31 (82)
Worse by 10% to <20%	3 (6)	6 (13)	1 (3)
Worse by ≥20%	7 (15)	5 (11)	6 (16)

^a Patients with 2 or more measurements at least a year apart. Values are presented as means (SD) in the top row and median (interquartile range) in the bottom row.

^b Patients with 2 or more measurements at least a year apart and within 5 years of the first measurement. Means (SD) in the top row and median (interquartile range) in the bottom row.

^c Change from first to second volume measurement if second volume is between 1 and 3 years after the first measurement. Means (SD) in the top row and median (interquartile range) in the bottom row.

^d Number of patients (%).

Cumulative MRI Utilization in Patients with Nonfunctional Adenoma

A total of 518 pituitary MRIs were performed in 95 patients with nonfunctional pituitary adenomas. Of these, 72 patients with microadenomas underwent 363 (70%) MRIs, including those obtained at initial presentation. Because all patients who ultimately required intervention had either symptoms or biochemical abnormalities on pituitary function testing, an individualized symptom- and laboratory-guided follow-up strategy after a baseline MRI at initial evaluation could have spared approximately 4 MRIs per patient with nonfunctional microadenoma over the course of their follow-up.

Discussion

In this retrospective cohort study of genetically confirmed and comprehensively phenotyped patients with MEN1, we evaluated the natural history and intervention rates of MEN1-related pituitary adenomas. Our findings reveal a cumulative proportion of 22% (17-27%) for nonfunctional pituitary adenomas in patients with MEN1 by age 40 years. Nevertheless, the rate of intervention for these adenomas was only 3.1% (2-4.7%) per 100 patient-years. All patients with nonfunctional adenomas who required intervention reported symptoms and/or developed biochemical abnormalities. A recent study of over 400 patients with sporadic pituitary microadenomas reported a minimal average growth rate of 0.016 mm/year (95% CI 0.037-0.069), suggesting that less frequent radiologic surveillance may be safe (21). Consistent with these findings, approximately 80% of patients in our cohort with nonfunctional adenomas demonstrated a median growth slope of 0 (IQR 0-8), indicating minimal tumor progression over time. These results call into question the benefit of routine pituitary MRI every 3 years in all patients with MEN1, as currently recommended.

Few studies have examined intervention rates for nonfunctional pituitary adenomas in MEN1 (8, 9, 14, 15). Improved resolution of MRIs over the years has enhanced our ability to detect these adenomas. However, the clinical benefit of such detection remains questionable. Similar to our observation of 3.1% (95% CI 2-4.7) rate of intervention per 100 patient-years for nonfunctional adenomas, a report from the Mayo Clinic described that in 52 patients with asymptomatic nonfunctional pituitary adenomas with 298 patient-years of follow-up, only 5 (10%) patients progressed to need surgery (9). The Dutch group reported that 91% (n = 35) of nonfunctioning pituitary adenomas did not require any intervention during a median follow-up of 6 years (14). There were 21/95 (22%) patients with nonfunctional adenomas who required an intervention in our cohort. However, all these patients reported symptoms or developed biochemical abnormalities preceding the need for an intervention. Thus, these patients could have been identified on history-taking and pituitary function testing for pituitary MRI. Similarly, a recent study of 84 MEN1-related pituitary adenomas (functional or nonfunctional) in a Spanish cohort identified a 7.2% (4/55 adenomas) risk of progression from microadenoma to invasive macroadenoma after 8 years of follow-up with prolactinomas showing the highest probability of progression (8). These findings now, replicated in multiple independent cohorts, underscore the lack of any benefit from radiological surveillance for pituitary adenomas in the vast majority of patients with MEN1. MEN1-related pituitary adenomas were originally described to have a more aggressive behavior and lower remission rate compared to sporadic adenomas (22-24). However, these recent observations in MEN1-related pituitary adenomas in multiple larger cohorts dismiss such conclusions. For context, 25% to 50% of sporadic nonfunctional adenomas exhibit tumor growth or visual field deterioration over 1 to 7 years of follow-up (25-29), and 20% of patients with nonfunctional macroadenomas on active surveillance required further intervention during a follow-up period of 7 years (30).

There are no reports of tumor volumetrics in MEN1-related pituitary adenomas. A study utilizing linear measurements in nonfunctional pituitary adenomas showed a median tumor increase of 2 mm over 5 years of follow-up (14). Similarly, a study of sporadic nonfunctional pituitary adenomas reported a mean annual growth rate of 1.0 mm/year for macroadenomas and 0.4 mm/year for microadenomas followed over a mean period of 4 years (31). Both these reports highlight the indolent nature of pituitary adenomas overall. Tumor volumetric analysis has greater power in detecting changes in tumor size and captures the lesion growth in 3 dimensions. Our observation of a relatively flat volume change over a mean 8.3 years of follow-up strengthens the evidence for the indolent nature of MEN1-related pituitary adenomas.

Multifocality is infrequent in MEN1-related pituitary adenomas but is more common than in patients with sporadic pituitary adenoma (6). There have been a few cases reported in MEN1 patients describing multiple synchronous functional adenomas. One case involved 2 distinct adenomas secreting prolactin and GH, respectively (32). Other reports have described histologically different adenomas, including somatotroph and null cell tumors, as well as corticotroph and gonadotroph adenomas (33, 34). In this study, only 1 patient (DK-381) was diagnosed with cosecretion of prolactin and GH.

Patients with MEN1 have an increased predisposition to prolactinomas. A study of Men^+/− mice showed that only pituitary lactotrophs and somatotrophs have proliferation rates that result in tumorigenesis over the murine lifespan (35). Of the 87 patients with functional adenomas, 78/87 (90%) in our cohort were prolactinomas. Prolactinomas in MEN1 were originally described to be more treatment-resistant and carry a higher risk of recurrence (22). Notably, 75/78 (96%) prolactinomas in our cohort had an intervention, predominantly using dopamine agonists, of which 22/75 (29%) patients needed surgery. However, only 3/75 (4%) patients required a second surgery, and 2 of those ultimately required radiation treatment. These findings are consistent with recent literature describing an appropriate response of MEN1-related prolactinomas to medical therapy (8, 14, 26). None of the patients in our cohort of MEN1 had pituitary carcinoma. Notably, MEN1-related pituitary carcinoma is extremely rare, with only a few cases reported in the literature (36-40).

The strengths of our study include (1) a longitudinal study of 320 well-characterized and genetically confirmed patients with MEN1, a rare heritable pituitary tumor syndrome; (2) a primary outcome of impact on patient management; (3) correlation of MRI findings and patient management to symptoms and biochemical abnormalities; (4) a central review of available MRIs for volume assessment; and (4) the use of tumor volumetrics to assess growth. Nevertheless, our study has limitations. First, this was a retrospective single-institution study and is thus subject to referral bias and unmeasured confounders. Second, since our study includes data from as far back as 1977, improvements in assays and imaging technology, as well as overall knowledge of the disease over time, can bias the approach to patient management. Third, the patients were evaluated at irregular intervals based on clinical judgment, shared decision-making, and feasibility of returning to the NIH Clinical Center for a follow-up appointment. Given that the outcome measure for the study is rate of intervention for pituitary adenoma, the frequency of MRI pituitary is not expected to have any impact on this outcome. Fourth, our cohort had a mean age at last follow-up of 47 ± 16 years. Given the relatively few children and elderly patients, our observations have limited applicability for these 2 groups of patients. Additionally, our sample size for macroadenomas is small and limits our ability to draw definitive conclusions about their natural history. Finally, it is important to note that the RECIST criteria are not validated for pituitary adenomas; however, we applied them based on their established use and validity in assessing other endocrine tumors.

To conclude, this longitudinal cohort study of MEN1-related pituitary adenomas shows that despite a high detection rate, most nonfunctional adenomas remained stable over long-term follow-up and did not require any intervention. The rate of intervention per 100 patient-years for nonfunctional adenomas was only 3.1%. All patients with nonfunctional adenomas who needed an intervention reported symptoms of headache or visual disturbances or developed abnormalities on pituitary function testing preceding the intervention. These findings suggest that radiological surveillance for nonfunctional pituitary adenomas in MEN1 offers minimal benefit over the assessment of symptoms and pituitary function testing in most patients. This is particularly relevant given the time, cost, limited availability, and potential risks associated with MRI, including patient anxiety, gadolinium contrast allergy, and renal injury. The current guidelines recommend a screening pituitary MRI at initial presentation in all patients with MEN1 (1). Based on our findings, we recommend surveillance for MEN1-related nonfunctional pituitary adenomas using an individualized approach that factors in the size of the adenoma (micro vs macro). Radiological surveillance may be necessary in select patients with microadenomas after an initial evaluation, based on tumor size and proximity to the optic chiasm. For patients with macroadenomas, radiological surveillance annually for the first 3 years as recommended for sporadic macroadenomas followed by its gradual spacing to 3 to 5 years in macroadenomas that demonstrate stability and do not have any suprasellar extension appears reasonable (41). Prospective studies are warranted to further validate these observations and identify appropriate ages for initiation and cessation of such surveillance.

Funding

This work was supported by the Intramural Research Programs of the National Institute of Diabetes and Digestive and Kidney Diseases (ZIADK043006).

Disclosures

All authors state that they have no conflicts of interest. The contributions of the NIH author(s) are considered Works of the United States Government. The findings and conclusions presented in this paper are those of the author(s) and do not necessarily reflect the views of the NIH or the US Department of Health and Human Services.

Data Availability

Some but not all data are available upon reasonable request to the corresponding author. Data sharing will require evaluation of the request by the local research ethics board and the signature of a data transfer agreement.