Atypical pituitary adenoma: a clinicopathologic case series · atypical pituitary adenoma treated at our institution be-tween 2007 and 2014. To our knowledge, this is one of the largest

CLINICAL ARTICLEJ Neurosurg 128:1058–1065, 2018

In 2004, the WHO released a new classification of pi-tuitary tumors that introduced atypical adenoma to the existing pituitary adenoma and carcinoma subtypes.

Atypical adenomas are defined by elevated mitotic index (> 2 mitoses per high-power field [HPF]), positive p53

staining (believed to be reflective of underlying mutation or pathway abnormality), and an MIB-1 proliferative in-dex ≥ 3% (Table 1).4 Although these criteria arose from older series suggesting that these features are markers of tumor aggressiveness,11,16,19,22,30–32 their clinical signifi-

ABBREVIATIONS ADC = apparent diffusion coefficient; HPF = high-power field; UCSF = University of California, San Francisco.SUBMITTED August 15, 2016. ACCEPTED December 21, 2016.INCLUDE WHEN CITING Published online June 9, 2017; DOI: 10.3171/2016.12.JNS162126.* Dr. Rutkowski and Mr. Alward contributed equally to this work.

Atypical pituitary adenoma: a clinicopathologic case series*Martin J. Rutkowski, MD,1 Ryan M. Alward, BS,1 Rebecca Chen, BA,1 Jeffrey Wagner, BA,1 Arman Jahangiri, BS,1 Derek G. Southwell, MD, PhD,1 Sandeep Kunwar, MD,1 Lewis Blevins, MD,1 Han Lee, MD,2 and Manish K. Aghi, MD, PhD1

Departments of 1Neurological Surgery and 2Neuropathology, California Center for Pituitary Disorders, University of California, San Francisco, California

OBJECTIVE In 2004, the WHO classified atypical pituitary adenoma as a distinct adenoma subtype. However, the clinical significance of this distinction remains undetermined. The authors sought to define patient characteristics, tumor features, and treatment outcomes associated with atypical pituitary adenoma.METHODS The authors reviewed records of patients who underwent resection of pituitary adenoma at the University of California, San Francisco, between 2007 and 2014. Per institutional protocol, adenomas exhibiting mitotic activity under-went evaluation for all 3 markers of atypicality (mitotic index, extensive p53 staining, and MIB-1 index ≥ 3%). Statistical analyses were performed using c2, Fisher’s exact test, t-test, log-rank, and logistic regression.RESULTS Between 2007 and 2014, 701 patients underwent resection for pituitary adenoma. Among these patients, 122 adenomas exhibited mitotic activity and therefore were evaluated for all 3 markers of atypicality, with 36 tumors (5%) proving to be atypical. There were 21 female patients (58%) and 15 male patients (42%) in the atypical cohort, and 313 female patients (47%) and 352 male patients (53%) in the nonatypical cohort (p = 0.231). The mean age of patients in the atypical cohort was 37 years (range 10–65 years), which was significantly lower than the mean age of 49 years (range 10–93 years) for patients in the nonatypical cohort (p < 0.001). The most common presenting symptoms for patients with atypical adenomas were headaches (42%) and visual changes (33%). Atypical adenomas were more likely to be func-tional (78%) than nonatypical adenomas (42%; p < 0.001). Functional atypical adenomas were significantly larger than functional nonatypical adenomas (mean diameter 2.2 vs 1.4 cm; p = 0.009), as were nonfunctional atypical adenomas compared with nonfunctional nonatypical adenomas (mean diameter 3.3 vs 2.3 cm; p = 0.01). Among the entire adeno-ma cohort, larger presenting tumor size was associated with cavernous sinus invasion (p < 0.001), and subtotal resection was associated with cavernous sinus invasion (p < 0.001) and larger size (p < 0.001) on binomial multivariate regression. The median time until recurrence was 56 months for atypical adenomas, 129 months for functional nonatypical adeno-mas, and 204 months for nonfunctional nonatypical adenomas (p < 0.001). Functional atypical adenomas recurred more frequently and significantly earlier than functional nonatypical adenomas (p < 0.001). When accounting for extent of resection, cavernous sinus invasion, size, age, sex, and functional subtype, atypicality remained a significant predictor of earlier recurrence among functional adenomas (p = 0.002).CONCLUSIONS When compared with nonatypical pituitary adenomas, atypical adenomas are more likely to present in younger patients at a larger size, are more often hormonally hypersecretory, and are associated with earlier recurrence. These features lend credence to atypical pituitary adenomas being a distinct clinical entity in addition to a discrete patho-logical diagnosis.https://thejns.org/doi/abs/10.3171/2016.12.JNS162126KEY WORDS atypical; pituitary; adenoma; recurrence; MIB-1; p53; aggressive; carcinoma; invasion; pituitary surgery

J Neurosurg Volume 128 • April 20181058 ©AANS 2018, except where prohibited by US copyright law

Unauthenticated | Downloaded 12/19/20 10:12 PM UTC

Atypical pituitary adenoma

J Neurosurg Volume 128 • April 2018 1059

cance in atypical pituitary adenoma remains largely un-investigated. Additionally, their impact on recurrence has not been investigated in a controlled fashion. This is an important oversight given that recurrence remains one of the most important end points for benign tumors such as pituitary adenomas, further necessitating new research on characteristics that delineate atypical adenomas as a dis-crete clinical entity. Indeed, the lack of studies to support or refute this classification scheme has created uncertainty about whether atypical pituitary adenoma should remain a distinct WHO classification.18 The evaluation of this clas-sification remains an important unanswered question giv-en the high prevalence of pituitary adenomas, estimated to be as high as 20% in the general population, and up to 15% of all intracranial tumors.5

The WHO’s efforts punctuate a philosophical evolu-tion in our understanding of pituitary adenoma biology. Formerly considered benign indolent tumors, there seems to be a subset of adenomas with relatively increased ag-gressiveness, which are defined by more than just tumor size (Fig. 1). However, it is unclear if the distinction be-tween nonatypical and atypical adenomas fully character-izes this subset of adenomas with greater aggressiveness.20 Given the paucity of data on atypical pituitary adenomas and their unclear clinical behavior, we compared atypical versus nonatypical pituitary adenomas at our large tertiary care center in an effort to better clarify features that define and delineate atypical tumors. We report patient demo-graphic data, tumor characteristics including size, hormone function, local invasion, histological markers, radiographic characteristics, and treatment history of 36 patients with atypical pituitary adenoma treated at our institution be-tween 2007 and 2014. To our knowledge, this is one of the largest series published to date, alongside a contempora-neous group of 665 nonatypical adenomas, in an effort to better understand features of atypical pituitary adenoma.

MethodsPatient Selection and Follow-Up

We reviewed the records of 701 patients who underwent surgical evaluation for pituitary adenoma at the California Center for Pituitary Disorders at the University of Califor-nia, San Francisco (UCSF), between 2007 and 2014. The UCSF institutional review board, known as the Committee on Human Research, reviewed and approved this study.

Among these 701 patients, 122 adenomas exhibited mi-totic activity on H & E staining and therefore underwent staining for mitotic index, p53, and MIB-1 proliferative in-dex per institutional protocol; 36 of these tumors met diag-nostic criteria for atypicality by showing positivity for all 3 features (5% of the total cohort, 30% of tumors stained for all 3 features of atypia). This included 5 patients who initially received surgery and postoperative management at other institutions; they subsequently demonstrated re-currence and underwent reoperation at UCSF with pathol-ogy-proven atypical pituitary adenoma. Demographic data were collected through medical record review, including primary treatment modality, adjuvant medical or radiation-based therapies, recurrence rates, and treatment strategies for recurrent tumors.

Per institutional practice, all patients underwent routine postoperative MRI within 3 months after primary surgery to assess extent of resection. Atypical tumors were fol-lowed with annual MRI scans, whereas nonatypical tu-mors were followed with annual scans for the first 3 years and with scans every 2 years thereafter.

Histological AssessmentPer the 2004 WHO designation, only patients with ad-

enomas meeting all 3 criteria for atypicality were included as having atypical pituitary adenomas: > 2 mitoses per 10 HPFs, positive p53 staining, and MIB-1 labeling index ≥ 3%. Histological confirmation of atypicality and hormone function was provided by 3 senior neuropathologists from the Department of Pathology and Laboratory Medicine at UCSF (see Acknowledgments).

Tumor CharacteristicsTumors were considered microadenomas when their

largest diameter was < 1 cm on MRI, and macroadenomas when ≥ 1 cm. Functionality (hormone hypersecretion) was determined on the basis of histological staining in con-junction with laboratory evidence of an abnormal preop-erative hormonal elevation.

Radiological CharacteristicsCavernous sinus invasion was assessed using Knosp cri-

teria,14 with Grades 3 and 4 (tumor invasion beyond a line tangential to the lateral margins of the cavernous internal carotid artery, and total internal carotid artery encasement, respectively) designating positive cavernous sinus invasion. Apparent diffusion coefficient (ADC) maps were reviewed on available preoperative MR images to test the hypoth-esis that atypical adenomas possess a unique radiographic ADC signature, on the basis of limited literature associat-ing ADC values with hypercellular pituitary tumors.1,23,29 Ovoid regions of interest 5–10 mm2 were drawn over pre-operative scans to assess ADC tumor values, and then nor-malized by dividing these values against similarly drawn regions of interest over white matter in the ipsilateral cen-trum semiovale. ADC ratios from nonatypical adenomas that did not meet any of the criteria for atypicality were used for comparison.

Biochemical EvaluationPre- and postoperative hormone testing data were avail-

TABLE 1. WHO criteria from 2004 for the diagnosis of nonatypical adenoma, atypical adenoma, and pituitary carcinoma

Pituitary Tumor Subtype

Pathological & Clinical Criteria for Diagnosis

Pituitary adenoma Monotonous histological features, infrequent mitotic figures, & Ki-67 index <3%


>2 mitoses per 10 HPFs, positive p53 immuno-staining, & Ki-67 index ≥3%

Pituitary carcinoma Disseminated disease &/or extraneural metastases


M. J. Rutkowski et al.

J Neurosurg Volume 128 • April 20181060

able for 280 of 303 patients with functional adenoma (92%), including 25 of 28 (89%) functional atypical adenomas and 255 of 275 (93%) functional nonatypical adenomas.

Assessment of RecurrenceRecurrence was determined radiologically, defined as

either MRI evidence of tumor reappearance following pri-mary gross-total resection, or as MRI evidence of tumor progression of known residual disease following primary subtotal resection. Time until recurrence was defined as the time from primary resection until the most recent MRI showing absent disease, stable residual disease, or recur-rent disease.

Statistical AnalysisAll statistical analysis was performed using SPSS

(IBM SPSS Statistics Developer 23.0), with significance defined as p < 0.05. c2 and Fisher’s exact tests were used for categorical variables, the Student t-test for continuous variables, and the log-rank test for recurrence data. Bino-mial logistic regression and Cox regression were used for multivariate analyses.

ResultsPatient Population

Between 2007 and 2014, 701 patients with pituitary adenoma underwent resection at our institution. Of these adenomas, 122 exhibited mitotic activity and therefore un-derwent evaluation for all 3 markers of atypicality, with 36 tumors (5%) proving to be atypical. Among the patients with nonatypical adenomas, there were 390 nonfunction-al adenomas (56%) and 275 functional adenomas (42%). There were 21 female patients (58%) and 15 male patients (42%) in the atypical cohort, and 313 female patients (47%) and 352 male patients (53%) in the nonatypical cohort (p = 0.231; Fisher’s exact test). The mean age of patients in the atypical cohort was 37 years (range 10–65 years), which was significantly lower than the mean age of 49 years (range 10–93 years) of patients in the nonatypical cohort (p < 0.0001, t-test). Data on preoperative hypopituitarism were available for 614 patients. Hypopituitarism was pres-ent in 44% of patients in the nonatypical cohort versus just 22% of patients in the atypical cohort on presentation (p =

0.009, Fisher’s exact test); more specifically, 53% of non-functional nonatypical adenomas and 33% of functional nonatypical adenomas showed preoperative hypopituita-rism. The most common presenting symptoms for patients with atypical adenomas were headaches (42%) and visual changes (33%).

Tumor Characteristics: Hormone Functionality and Staining

Atypical adenomas were more likely to be functional (28 of 36 [78%]) than nonatypical adenomas (275 of 665 [42%]) (p < 0.0001; Fisher’s exact test). Among the 8 atyp-ical adenomas that were clinically nonfunctional, there were 6 null cell tumors (16.7%), 1 silent gonadotroph (3%), and 1 silent lactotroph (3%). Among the functional atypical adenomas, there were 16 lactotrophs (44%) causing hyper-prolactinemia, 6 corticotrophs (16.7%) causing Cushing’s disease, 4 somatotrophs (11%) causing acromegaly, and 2 lactotroph/somatotrophs (5.6%) causing mixed hyper-prolactinemia/acromegaly, respectively. Functional non-atypical adenomas exhibited a similar breakdown, with 134 lactotrophs (48.7%) causing hyperprolactinemia, 77 somatotrophs (28%) causing acromegaly, 54 corticotrophs (19.6%) causing Cushing’s disease, 9 mixed lactotroph/somatotrophs (3.3%) causing mixed hyperprolactinemia/acromegaly, respectively, and 1 thyrotroph (< 1%) causing hyperthyroidism.

There was a total of 303 functional tumors. Overall, there were 150 lactotrophs (49.5%), 60 corticotrophs (19.8%), 81 somatotrophs (26.7%), 11 mixed lactotroph/somatotrophs (3.6%), and 1 thyrotroph (< 1%). Hyperpro-lactinemia (p = 0.433, Fisher’s exact test), acromegaly (p = 0.177, Fisher’s exact test), and hypercortisolemia (p = 0.805, Fisher’s exact test) were not associated with atypi-cal versus nonatypical functional adenomas.

Atypical Versus Nonatypical Pituitary Adenoma: Preoperative Imaging

To better characterize clinical differences between atypical and nonatypical pituitary adenoma, we compared adenomas on the basis of size, invasion, and recurrence. Among the entire adenoma cohort, adenomas ranged from 0.2 to 7.3 cm, with a mean size of 1.9 cm. There

FIG. 1. Three T1-weighted postcontrast MR sequences in a 51-year-old man with recurrent atypical pituitary adenoma. These sagittal (A), coronal (B), and axial (C) images demonstrate a 7.3-cm recurrence of a previously resected gonadotrophic pituitary adenoma. Pathology on repeat resection demonstrated > 2 mitoses per 10 HPF, extensive p53 immunohistochemical positivity, and elevated MIB-1 index ≥ 3%, consistent with atypical pituitary adenoma. Note diffuse local invasion of the sphenoid sinus, cli-vus, cavernous sinus, and beyond the suprasellar cistern, with mild mass effect on the right uncus and anterior temporal lobe. The tumor demonstrates encasement of the right internal carotid artery with mass effect on the left internal carotid and basilar arteries.




was a significant difference in the size of atypical ver-sus nonatypical pituitary adenomas on presentation; the mean atypical tumor diameter was 2.4 versus 1.9 cm for nonatypical tumors (p = 0.021, t-test). Macroadenomas were seen in 8 of 8 (100%) nonfunctional atypical adeno-mas, 23 of 28 (82%) functional atypical adenomas, 325 of 342 (95%) nonfunctional nonatypical adenomas, and 166 of 275 (60%) functional nonatypical adenomas. The increased size in atypical adenomas compared with non-atypical adenomas held true when analyzing the function-al and nonfunctional cohorts. Functional atypical adeno-mas were significantly larger than functional nonatypical adenomas (mean diameter 2.2 vs 1.4 cm; p = 0.009, t-test), as were nonfunctional atypical adenomas compared with nonfunctional nonatypical adenomas (mean diameter 3.3 vs 2.3 cm; p = 0.01, t-test).

Cavernous sinus invasion was encountered significant-ly more often in atypical adenomas than nonatypical ad-enomas (44% vs 27%; p = 0.047, Fisher’s exact test), even when substratified by functional atypical versus function-al nonatypical adenomas (48% vs 24%; p = 0.012, Fisher’s exact test). Larger atypical and nonatypical tumors were more often associated with cavernous sinus invasion (p = 0.024 and p < 0.001, respectively, t-test). When controlling for size, binary logistic regression revealed that atypicality lost significance as a predictor of cavernous sinus invasion (p = 0.150). There was no relationship between functional-ity and invasion (p = 0.543, Fisher’s exact test). Among the 122 adenomas that were stained for all 3 features of atypi-cality, there was no relationship between cavernous sinus invasion and MIB-1 labeling index (p = 0.730, t-test), p53 staining (p = 0.897, t-test), and mitotic index (p = 0.568, t-test).

ADC ratios were available and calculated for 28 non-atypical adenomas and 25 atypical adenomas. There was no significant difference in ADC ratio between atypical and nonatypical adenomas (p = 0.591, t-test), including when stratified by functional atypical versus functional nonatypical adenomas (p = 0.114, t-test), or by nonfunc-tional atypical versus nonfunctional nonatypical adeno-mas (p = 0.830, t-test).

Atypical Versus Nonatypical Pituitary Adenoma: Extent of Resection

There were gross-total resections in 5 of 8 (63%) non-functional atypical adenomas, 13 of 28 (46%) functional atypical adenomas, 185 of 319 (58%) nonfunctional non-atypical adenomas, and 159 of 237 (67%) functional non-atypical adenomas. Subtotal resection was more com-monly associated with functional atypical adenoma than functional nonatypical adenomas (54% vs 33%; p = 0.037, Fisher’s exact test), invasive tumors (p < 0.001, Fisher’s ex-act test), and large tumors (2.6 vs 1.7 cm; p < 0.001, t-test). Cavernous sinus invasion was more often associated with subtotal resection (p < 0.001, Fisher’s exact test). Whereas cavernous sinus invasion (p < 0.001; 95% CI 4.9–12.2) and larger size (p < 0.001; 95% CI 1.4–2.1) remained signifi-cant predictors of subtotal resection on binomial multi-variate regression, atypicality dropped out of the stepwise model (p = 0.394).

Biochemical Remission for Functional Atypical Versus Nonatypical Adenomas

Biochemical remission following primary resection was seen in 15 of 25 (60%) functional atypical adenomas and 192 of 255 (75%) functional nonatypical adenomas (p = 0.101, Fisher’s exact test). Biochemical remission was also strongly associated with gross-total versus subtotal resection (p < 0.0001, Fisher’s exact test) and smaller tu-mor size (1.2 vs 2.4 cm; p < 0.0001). Given the trend toward significance among atypical versus nonatypical functional adenomas, each functional subtype (lactotrophs, somato-trophs, and corticotrophs) was plugged into a multivari-ate model, including tumor size and extent of resection. Following binomial logistic regression, only tumor size (p = 0.018; 95% CI 0.50–0.94) and extent of resection (p < 0.0001; 95% CI 4.2–18.9) remained significant predic-tors of biochemical remission. There was no significant association between atypicality versus nonatypicality and rates of biochemical remission.

Recurrence of Atypical Versus Nonatypical AdenomasThere were 7 recurrences (19%) among the atypical ad-

enomas, 33 recurrences (8.5%) among the nonfunctional nonatypical adenomas, and 33 recurrences (12%) among the functional nonatypical adenomas. When comparing the time until recurrence among atypical, functional non-atypical, and nonfunctional nonatypical adenomas, there was a progressive increase in time until recurrence: the median time until recurrence was 56 months for atypical adenomas, 129 months for functional nonatypical adeno-mas, and 204 months for nonfunctional nonatypical ad-enomas (Fig. 2; p < 0.001, log-rank). Although there were too few nonfunctional atypical adenomas to compare with nonfunctional nonatypical adenomas, functional atypical adenomas recurred significantly earlier than functional

FIG. 2. Kaplan-Meier curve representing time until recurrence as strati-fied by atypical, functional nonatypical, and nonfunctional nonatypical pathology. There was a significant difference between time until recur-rence among the 3 cohorts (p < 0.0001, log-rank). APA = atypical pitu-itary adenoma; FPA = functional nonatypical pituitary adenoma; NFPA = nonfunctional nonatypical pituitary adenoma.




nonatypical adenomas (Fig. 3; p < 0.001, log-rank). When accounting for extent of resection, cavernous sinus inva-sion, size, age, sex, and functional subtype, atypicality remained a significant predictor of earlier functional ad-enoma recurrence on Cox regression (p = 0.002).

The MIB-1 index ranged from 3% to 20%, with a mean of 4.1%. When examining the 122 adenomas that were stained for all 3 features of atypicality, MIB-1 index was the single feature that nearly achieved significance when examining time until recurrence (p = 0.050001, log-rank) (Fig. 4). Mitotic index and p53 staining were not signifi-cantly associated with time until recurrence (p = 0.913 and p = 0.512, respectively, log-rank).

Management of Recurrent DiseaseSix of 7 recurrent atypical pituitary adenomas under-

went repeat transsphenoidal resection. There were 5 sub-total resections and 1 gross-total resection (Table 2). Ad-juvant therapy was further used for 3 patients. One patient with recurrent Cushing’s disease underwent repeat resec-tion and was treated with adjuvant pasireotide to improve biochemical control; another patient with recurrent Cush-ing’s disease underwent repeat resection twice, received adjuvant radiosurgery, and then had bilateral adrenalec-tomy for refractory hypercortisolemia. One patient with a recurrent lactotroph underwent repeat resection twice and was given cabergoline for improved biochemical control.

Adjuvant therapy and/or salvage therapy for recurrent nonatypical tumors depended on functionality. For func-tional adenomas, intrasellar recurrence or accessible cav-ernous sinus and/or suprasellar disease was treated with repeat surgery, whereas microscopic residual disease with persistent hormonal hypersecretion was treated with medical therapy. Surgically inaccessible recurrences were treated with medical and radiation-based adjuvant thera-py. Nonfunctional, nonatypical recurrences were treated

with repeat surgery for accessible disease and radiation-based adjuvant therapy for invasive disease.

DiscussionTo our knowledge, we report here one of the larg-

est case series of patients with histologically confirmed atypical pituitary adenoma following resection. Due to the relatively recent differentiation of pituitary tumor sub-types (Table 1) by the WHO in 2004, the limited num-ber of studies that led to these criteria,11,16,19,22,30–32 and the limited number of subsequent studies,2,3,12,20,25,33,35,37,38 the clinical consequences of atypicality have yet to be fully elucidated. This makes a better understanding of their be-havior paramount to validating the designation of atypical adenomas as a separate pathological entity.

The frequency of atypical pituitary adenoma has been reported to be as high as 15%, with the largest and most recent studies reporting rates of 2.7%,24 8.9%,35 11.6%,3 14.7%,27 and 14.8%37 (Table 3). Our series indicates a relatively lower rate of 5%. Our institutional protocol of screening for all 3 features of atypicality among tumors with mitotic activity may have caused our atypicality inci-dence to be relatively lower than it would have been with complete screening for all 3 features in all cases.

We note a median age of 37 years and a slight female pre-dominance (58%) among patients in our case series. Both of these values are lower than those in another large series of atypical pituitary adenomas, in which the mean age was 48 years with a female predominance of 67%.37 Atypical pituitary adenoma seemed to present more often in young-er patients relative to nonatypical tumors, and more often as large tumors than did nonatypical tumors, for both the functional and nonfunctional subtypes. Both observations probably reflect the more aggressive and proliferative biol-ogy of atypical tumors. Indeed, in another large series of 50 patients with atypical adenoma, Del Basso De Caro et

FIG. 3. Kaplan-Meier curve representing time until recurrence as strati-fied by functional atypical and functional nonatypical pathology. There was a significant difference between time until recurrence (p < 0.001, log-rank). fAPA = functional atypical pituitary adenoma; FPA = functional nonatypical pituitary adenoma.

FIG. 4. Kaplan-Meier curve representing time until recurrence as strati-fied by MIB-1 labeling index. There was a significant difference between time until recurrence among tumors with an elevated MIB-1 index (36 adenomas) compared with those without an elevated index (86 adeno-mas; p = 0.050001).




al. reported a relatively lower age for patients with atypical adenoma (42 years), a high proportion of macroadenomas (82%), and relatively high rates of cavernous sinus invasion (42%);3 these rates are similar to those in our series and are further suggestive of more aggressive behavior in atypical adenomas. Zada et al.37 reported that 94% of patients in their cohort presented with macroadenomas, and Yildirim et al.35 reported that all 13 of their patients were noted to have macroadenomas (Table 3).

Interestingly, the hormone profile among atypical ad-enomas shows some variability. Whereas we report that only 22% of atypical tumors were nonfunctional, they were among the most common subtypes encountered in multiple other studies, along with tumors that stained positive for growth hormone and adrenocorticotropic hor-mone.25,35,37 In contrast, lactotrophs were most common (44%) in our study. Our finding that atypical adenomas are more significantly associated with hormone hypersecre-tion has not been previously reported.

We found a significant association between atypical pathology and unilateral or bilateral cavernous sinus inva-sion on univariate analysis, which lost significance when tumor size was included in logistic regression. Thus, al-though atypical tumors may not be more locally invasive than nonatypical tumors, their increased capacity for pro-liferation may in fact be responsible for their association with cavernous sinus invasion. The lack of a relationship between atypicality and invasiveness in our study con-trasts with other studies in which atypicality has been as-sociated with invasiveness. Zada et al.37 reported that 83% of atypical tumors showed local invasion, and Yildirim et al.35 reported 100%. Although invasion has been posited as a feature of atypicality, it is important to note that non-atypical pituitary adenomas can demonstrate invasion, and not all atypical adenomas invade locally.5

A radiographic distinction between atypical and non-atypical tumors could be valuable as a means of influenc-ing treatment decision making for asymptomatic tumors at the time of initial evaluation. Whereas contemporaneous work performed at our center29 demonstrated an associa-tion between higher MIB-1 index and lower ADC values, our study found that the correlation between ADC signa-ture and atypicality was insignificant. Lower ADC values are thought to reflect intratumoral hypercellularity and

thus may act as a surrogate for more proliferative tumors reflected by higher MIB-1 indices. However, our finding underscores the challenge of identifying a radiographic biomarker that correlates with the 3 pathologic features defining atypicality.

Our finding that hypopituitarism at presentation was more often associated with nonatypical tumors is unex-pected. It is possible that hypopituitarism is more a result of duration of mass effect rather than severity of mass ef-fect, such that the larger size and younger patient age asso-ciated with atypical adenomas lead to less glandular dam-age than chronic glandular compression caused by more insidious growth of nonatypical tumors.

The lack of a correlation between atypicality and cavern-ous sinus invasion in multivariate analysis probably contrib-utes to our finding that atypicality similarly did not affect extent of resection and biochemical remission for functional adenomas. Nevertheless, our overall recurrence rate among atypical adenomas was 19% with a median of 56 months, which was significantly earlier than among functional and nonfunctional nonatypical adenomas. This remained true even when controlling for extent of resection, cavernous sinus invasion, size, age, sex, and functional subtype, and further supports the concept that atypical adenomas are distinct from nonatypical tumors on the basis of clinical characteristics in addition to histological differences. These differences in recurrence rates may reflect the fact that mi-croscopic residual disease is likely to reach radiographic or biochemical significance more quickly for an atypical ad-enoma because atypical cells are more proliferative.

Our recurrence rate (19%) seems slightly lower and more delayed than those reported in other case series; Yildirim et al.35 reported 38.4% recurrence at 41 months postoperative, and Zada et al.37 reported a recurrence rate of 22% at 4 years postoperative. Notably, in the latter study, there was no significant difference in the recurrence rates between their series of atypical and typical adenomas, un-like the findings in our series. Our analysis demonstrating earlier recurrence as a function of MIB-1 index nicely par-allels the work of Miermeister et al.,20 who showed that an index > 4% was the most reliable diagnostic indicator of atypical pituitary adenoma. Given our finding that atypi-cality results in larger tumors with earlier recurrences, this marker may be a particularly important prognostic feature.

TABLE 2. Clinical characteristics and courses for patients who underwent repeat surgery for recurrent atypical pituitary adenoma

Age (yrs), Sex

Presenting Signs & Symptoms

Hormone Positivity

Prior Treatment(s)

Recurrence-Free Survival (mos)

Invasion on Recurrence

Size on Recurrence

(cm)

UCSF Treatment for Recurrence

51, M Blurry vision, memory dysfunction Gonadotroph Resection 106 Yes 7.3 STR28, F Hypercortisolism, headaches,

blurry visionCorticotroph Resection 11 Yes 4.1 STR

31, F Hypercortisolism Corticotroph Resection + pasireotide 39 No 0.9 STR59, F Hypercortisolism, visual changes Corticotroph Resection + RT, adrenalec-

tomy, repeat surgery ×169 Yes 2.7 STR

44, F Asymptomatic w/ enlarging tumor on MRI

Lactotroph Resection + cabergoline, repeat surgery ×1

72 No 2.0 GTR

58, M Headaches Corticotroph Resection 30 Yes 1.7 STR

GTR = gross-total resection; RT = radiotherapy; STR = subtotal resection.




The criteria designated by the WHO for atypical ad-enomas emerged from the work of multiple investigators examining biomarkers for their correlation with tumor be-havior and prognosis. p53 staining tends to be strongly pos-itive in large, invasive, and aggressive adenomas; 1 study reported a correlation between p53 positivity and tumor behavior, with expression levels at 0% among noninvasive adenomas, 15.2% among invasive adenomas, and 100% among pituitary carcinomas.31 This association between p53 positivity and invasion and recurrence has been cor-roborated15,21,34 and disputed9,27,28,31 by the literature. MIB-1 labeling remains another equivocal marker of pituitary tumor behavior, due in no small part to different reporting standards, and the literature similarly reports disparate efficacy in predicting invasion and recurrence.6,7,9,10,13,17, 20,

26,36 Miermeister et al. recently reexamined a large Ger-man registry of patients with pituitary tumors only to find that MIB-1 index > 4% was most reliably associated with atypical diagnosis on receiver operating characteris-tic analysis.20 Beyond these markers already used in the definition of atypical adenomas, others have investigated the utility of enzymes such as matrix metalloproteinases (MMPs; including MMP2 and MMP9) and growth fac-tors and their receptors (epidermal growth factor/receptor, vascular endothelial growth factor/receptor, and fibroblast growth factor/receptor) as potential biomarkers of aggres-sive adenoma biology.5,8

Limitations of the StudyIt is important to note that 5 of 7 recurrences in the

atypical pituitary adenoma cohort occurred in patients who received their initial resection at an outside institu-tion. Furthermore, our study is retrospective. Finally, the incidence of atypical pituitary adenoma in our series may be difficult to interpret given the lack of uniformity in staining for mitotic index, p53, and MIB-1 index among all patients with pituitary adenoma treated at our center since the 2004 WHO criteria were determined.

ConclusionsDiagnosis of atypical pituitary adenoma necessitates

the presence of 3 pathological criteria, which include > 2 mitoses per HPF, p53 immunostaining, and MIB-1 index ≥ 3%. When compared with nonatypical pituitary adenoma, atypical adenomas are more likely to present in younger

patients at a larger size, are more often hormonally hy-persecretory, and are associated with earlier recurrence. These findings represent the first and most comprehensive side-by-side comparison of atypical versus nonatypical ad-enomas, and, for the first time, lend credence to the WHO designation of atypical adenoma as a unique pituitary tu-mor subtype.

AcknowledgmentsWe acknowledge the assistance of Tarik Tihan, MD, PhD, Arie

Perry, MD, and Andrew Bollen, MD, from the Neuropathology Division of the Department of Pathology and Laboratory Medicine at UCSF.

Mr. Alward received a UCSF Dean’s Fellowship for completion of these studies.

References 1. Boxerman JL, Rogg JM, Donahue JE, Machan JT, Gold-

man MA, Doberstein CE: Preoperative MRI evaluation of pituitary macroadenoma: imaging features predictive of successful transsphenoidal surgery. AJR Am J Roentgenol 195:720–728, 2010

2. Chiloiro S, Doglietto F, Trapasso B, Iacovazzo D, Giampietro A, Di Nardo F, et al: Typical and atypical pituitary adenomas: a single-center analysis of outcome and prognosis. Neuroen-docrinology 101:143–150, 2015

3. Del Basso De Caro M, Solari D, Pagliuca F, Villa A, Gua-dagno E, Cavallo LM, et al: Atypical pituitary adenomas: clinical characteristics and role of ki-67 and p53 in prognos-tic and therapeutic evaluation. A series of 50 patients. Neuro-surg Rev 40:105–114, 2017

4. DeLellis RA, Lloyd RV, Heitz PU, Eng C (eds): World Health Organization Classification of Tumours, Vol. 8. Pathology and Genetics. Tumours of Endocrine Organs. Lyon, France: IARC, 2004

5. Di Ieva A, Rotondo F, Syro LV, Cusimano MD, Kovacs K: Aggressive pituitary adenomas—diagnosis and emerging treatments. Nat Rev Endocrinol 10:423–435, 2014

6. Fusco A, Zatelli MC, Bianchi A, Cimino V, Tilaro L, Veltri F, et al: Prognostic significance of the Ki-67 labeling index in growth hormone-secreting pituitary adenomas. J Clin Endo-crinol Metab 93:2746–2750, 2008

7. Gejman R, Swearingen B, Hedley-Whyte ET: Role of Ki-67 proliferation index and p53 expression in predicting progres-sion of pituitary adenomas. Hum Pathol 39:758–766, 2008

8. Heaney A: Management of aggressive pituitary adenomas and pituitary carcinomas. J Neurooncol 117:459–468, 2014

9. Hentschel SJ, McCutcheon E, Moore W, Durity FA: P53 and MIB-1 immunohistochemistry as predictors of the clinical

TABLE 3. Summary of major published studies on atypical pituitary adenoma

Authors & Year Sex

Mean Age (yrs)

Incidence (%)

MA (%)

Functional (%)

CSI (%)

Mean MIB-1 Index & Range (%)

Recurrence Rate (%)

Time Until Recurrence (mos)

Present study 21 F, 15 M 37 5 80 78 44 4.1 (3–20) 19 56Del Basso De Caro et al., 2017 24 F, 26 M 42 11.6 82 54 42 5.6 (3.5–22.5) 12 18Saeger et al., 2016 47 F, 51 M 46 2.7 NA NA 47 9 NA NAYildirim et al., 2013 3 F, 10 M 53 8.9 100 31 69 4.7 (3–10) 38 41Zada et al., 2011 12 F, 6 M 44 14.8 94 50 33 7 (3–20) 22 48

Among the larger case series represented, atypical adenoma incidence ranged from 2.7% to 14.8%; incidence of cavernous sinus invasion was 33%–69%, with a recur-rence rate of 12%–38%. Most atypical adenomas were reported as macroadenomas, as in our series, although rates of hormone functionality were lower in other series (31%–54% vs 78% in our series).CSI = cavernous sinus invasion; MA = macroadenoma; NA = not available.




behavior of nonfunctioning pituitary adenomas. Can J Neu-rol Sci 30:215–219, 2003

10. Honegger J, Prettin C, Feuerhake F, Petrick M, Schulte-Mönting J, Reincke M: Expression of Ki-67 antigen in nonfunctioning pituitary adenomas: correlation with growth velocity and invasiveness. J Neurosurg 99:674–679, 2003

11. Jaffrain-Rea ML, Di Stefano D, Minniti G, Esposito V, Bultrini A, Ferretti E, et al: A critical reappraisal of MIB-1 labelling index significance in a large series of pituitary tumours: secreting versus non-secreting adenomas. Endocr Relat Cancer 9:103–113, 2002

12. Kim JS, Lee YS, Jung MJ, Hong YK: The predictive value of pathologic features in pituitary adenoma and correlation with pituitary adenoma recurrence. J Pathol Transl Med 50:419–425, 2016

13. Knosp E, Kitz K, Perneczky A: Proliferation activity in pitu-itary adenomas: measurement by monoclonal antibody Ki-67. Neurosurgery 25:927–930, 1989

14. Knosp E, Steiner E, Kitz K, Matula C: Pituitary adenomas with invasion of the cavernous sinus space: a magnetic reso-nance imaging classification compared with surgical findings. Neurosurgery 33:610–618, 1993

15. Kontogeorgos G: Predictive markers of pituitary adenoma behavior. Neuroendocrinology 83:179–188, 2006

16. Losa M, Barzaghi RL, Mortini P, Franzin A, Mangili F, Ter-reni MR, et al: Determination of the proliferation and apop-totic index in adrenocorticotropin-secreting pituitary tumors: comparison between micro- and macroadenomas. Am J Pathol 156:245–251, 2000

17. Losa M, Franzin A, Mangili F, Terreni MR, Barzaghi R, Veg-lia F, et al: Proliferation index of nonfunctioning pituitary adenomas: correlations with clinical characteristics and long-term follow-up results. Neurosurgery 47:1313–1319, 2000

18. Louis DN, Perry A, Reifenberger G, von Deimling A, Fig-arella-Branger D, Cavenee WK, et al: The 2016 World Health Organization Classification of Tumors of the Central Nervous System: a summary. Acta Neuropathol 131:803–820, 2016

19. Mastronardi L, Guiducci A, Spera C, Puzzilli F, Liberati F, Maira G: Ki-67 labelling index and invasiveness among ante-rior pituitary adenomas: analysis of 103 cases using the MIB-1 monoclonal antibody. J Clin Pathol 52:107–111, 1999

20. Miermeister CP, Petersenn S, Buchfelder M, Fahlbusch R, Lüdecke DK, Hölsken A, et al: Histological criteria for atypi-cal pituitary adenomas—data from the German pituitary adenoma registry suggests modifications. Acta Neuropathol Commun 3:50, 2015 (Erratum in Acta Neuropathol Com-mun 4:21, 2016)

21. Ozer E, Canda MS, Ulukus C, Guray M, Erbayraktar S: Expression of Bcl-2, Bax and p53 proteins in pituitary ad-enomas: an immunohistochemical study. Tumori 89:54–59, 2003

22. Pernicone PJ, Scheithauer BW, Sebo TJ, Kovacs KT, Horvath E, Young WF Jr, et al: Pituitary carcinoma: a clinicopatho-logic study of 15 cases. Cancer 79:804–812, 1997

23. Pierallini A, Caramia F, Falcone C, Tinelli E, Paonessa A, Ciddio AB, et al: Pituitary macroadenomas: preoperative evaluation of consistency with diffusion-weighted MR imag-ing—initial experience. Radiology 239:223–231, 2006

24. Saeger W, Honegger J, Theodoropoulou M, Knappe UJ, Schöfl C, Petersenn S, et al: Clinical impact of the current WHO classification of pituitary adenomas. Endocr Pathol 27:104–114, 2016

25. Saeger W, Lüdecke DK, Buchfelder M, Fahlbusch R, Quabbe HJ, Petersenn S: Pathohistological classification of pituitary tumors: 10 years of experience with the German Pituitary Tumor Registry. Eur J Endocrinol 156:203–216, 2007

26. Salehi F, Agur A, Scheithauer BW, Kovacs K, Lloyd RV,

Cusimano M: Ki-67 in pituitary neoplasms: a review—part I. Neurosurgery 65:429–437, 2009

27. Scheithauer BW, Gaffey TA, Lloyd RV, Sebo TJ, Kovacs KT, Horvath E, et al: Pathobiology of pituitary adenomas and carcinomas. Neurosurgery 59:341–353, 2006

28. Suliman M, Royds J, Cullen D, Timperley W, Powell T, Bat-tersby R, et al: Mdm2 and the p53 pathway in human pitu-itary adenomas. Clin Endocrinol (Oxf) 54:317–325, 2001

29. Tamrazi B, Pekmezci M, Aboian M, Tihan T, Glastonbury CM: Apparent diffusion coefficient and pituitary macroad-enomas: pre-operative assessment of tumor atypia. Pituitary [epub ahead of print], 2016

30. Thapar K, Kovacs K, Scheithauer BW, Stefaneanu L, Horvath E, Pernicone PJ, et al: Proliferative activity and invasiveness among pituitary adenomas and carcinomas: an analysis using the MIB-1 antibody. Neurosurgery 38:99–107, 1996

31. Thapar K, Scheithauer BW, Kovacs K, Pernicone PJ, Laws ER Jr: p53 expression in pituitary adenomas and carcinomas: correlation with invasiveness and tumor growth fractions. Neurosurgery 38:765–771, 1996

32. Thapar K, Yamada Y, Scheithauer B, Kovacs K, Yamada S, Stefaneanu L: Assessment of mitotic activity in pituitary ad-enomas and carcinomas. Endocr Pathol 7:215–221, 1996

33. Tortosa F, Webb SM: Atypical pituitary adenomas: 10 years of experience in a reference centre in Portugal. Neurologia 31:97–105, 2016

34. Wierinckx A, Auger C, Devauchelle P, Reynaud A, Cheval-lier P, Jan M, et al: A diagnostic marker set for invasion, pro-liferation, and aggressiveness of prolactin pituitary tumors. Endocr Relat Cancer 14:887–900, 2007

35. Yildirim AE, Divanlioglu D, Nacar OA, Dursun E, Sahinoglu M, Unal T, et al: Incidence, hormonal distribution and post-operative follow up of atypical pituitary adenomas. Turk Neurosurg 23:226–231, 2013

36. Yokoyama S, Hirano H, Moroki K, Goto M, Imamura S, Ku-ratsu JI: Are nonfunctioning pituitary adenomas extending into the cavernous sinus aggressive and/or invasive? Neuro-surgery 49:857–863, 2001

37. Zada G, Woodmansee WW, Ramkissoon S, Amadio J, Nose V, Laws ER Jr: Atypical pituitary adenomas: incidence, clini-cal characteristics, and implications. J Neurosurg 114:336–344, 2011

38. Zaidi HA, Cote DJ, Dunn IF, Laws ER Jr: Predictors of ag-gressive clinical phenotype among immunohistochemically confirmed atypical adenomas. J Clin Neurosci 34:246–251, 2016

DisclosuresThe authors report no conflict of interest concerning the materi-als or methods used in this study or the findings specified in this paper.

Author ContributionsConception and design: Rutkowski, Alward, Kunwar, Blevins, Aghi. Acquisition of data: Rutkowski, Alward, Chen, Wagner. Analysis and interpretation of data: Rutkowski, Kunwar, Aghi. Drafting the article: Rutkowski, Alward, Aghi. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Rutkowski. Statistical analysis: Rutkowski. Administrative/technical/material support: Aghi.

CorrespondenceMartin J. Rutkowski, Department of Neurological Surgery, Uni-versity of California, San Francisco, 505 Parnassus Ave., M-779, San Francisco, CA 94143. email: [email protected].


Documents

Atypical pituitary adenoma: a clinicopathologic case series · atypical pituitary adenoma treated at our institution be-tween 2007 and 2014. To our knowledge, this is one of the largest