Brain Tumor Pathology

https://doi.org/10.1007/s10014-017-0307-7

INSTRUCTIONAL LECTURE

New WHO classification of pituitary adenomas (4th edition):

assessment of pituitary transcription factors and the prognostic
histological factors
Hiroshi Nishioka1,3 · Naoko Inoshita2,3

Received: 20 December 2017 / Accepted: 26 December 2017

© The Japan Society of Brain Tumor Pathology 2018

Abstract
WHO classification of pituitary adenomas was revised in 2017. The two major and significant changes are discussed. (1) The
new classification focuses on adenohypophysial-cell lineage for the designation of adenomas, and thus, assessment of pitui-
tary transcription factors is recommended. Its appropriate use has a complementary role in obtaining an accurate diagnosis,
particularly in hormone-negative adenomas. Subclassification of nonfunctioning adenomas was revised accordingly and,
consequently, null cell adenomas became quite rare. (2) “Atypical adenoma”, a previous category, was eliminated due to the
poor reproducibility and predictive value. Assessment of tumor proliferation marker and other clinical parameters such as
invasion are recommended to predict aggressiveness. “High-risk adenomas” are those with rapid growth, radiological inva-
sion, and a high Ki-67 proliferation index, whereas some special adenoma subtypes commonly show aggressive behavior.

Keywords  Aggressive adenomas · Nonfunctioning adenomas · Pituitary adenomas · Transcription factor · WHO
classification

Introduction “aggressive” adenomas are discussed, whereas a previous

category “atypical” adenoma was eliminated. The benefits
“WHO classification of tumours of endocrine organs” was and misleading of the new classification are also discussed.
revised in 2017 (4th edition). 13 years have passed since the
last 3rd edition in 2004. There are several changes in the new
classification of pituitary adenomas according to the recent Pituitary transcription factors
new knowledge including tumor development and prognosis
[1–3]. In this article, the two major and significant changes The recent advances in molecular biology have clarified
are focused. (1) For the new classification, a novel approach the cytodifferentiation pathways involved in the develop-
according to adenohypophysial-cell lineages was introduced. ment of adenohypophysial cells. During development,
Assessment of pituitary transcription factors is required in an organized, complex process of cell differentiation is
some adenomas for the accurate subclassification, particu- orchestrated by specific transcription factors. The pituitary
larly for nonfunctioning adenomas. (2) Clinical and patho- transcription factors also have a role in determining the
logical prognostic parameters for predicting “high-risk” and cytodifferentiation and hormone production of pituitary
adenomas and, thus, can serve as diagnostic markers [4–6].
The new WHO classification proposed the application of
* Hiroshi Nishioka transcription factors immunohistochemistry to determine
nishioka@tokyo‑med.ac.jp
adenohypophysial cell lineages for the accurate subclas-
1
Department of Hypothalamic and Pituitary Surgery, sification of pituitary adenomas [1–3]. There are three
Toranomon Hospital, 2‑2‑2 Toranomon, Minatoku, main pathways of adenohypophysial cell differentiation
Tokyo 105‑8470, Japan and the determined transcription factors (Fig. 1): (1) cor-
2
Department of Pathology, Toranomon Hospital, Tokyo, Japan ticotrophs determined by the t-box pituitary transcription
3
Okinaka Memorial Institute for Medical Research, Tokyo, factor (Tpit), (2) somatotrophs/lactotrophs/mammosoma-
Japan totrophs/thyrotrophs determined by pituitary transcription

13
Vol.:(0123456789)
 Brain Tumor Pathology

Fig. 1  Adenohypophysial cell
lineage and the transcription
factors

factor 1 (Pit-1), and (3) gonadotrophs determined by ster- Gonadotroph adenomas

oidogenic Factor-1 (SF-1) and/or GATA-2 in the presence
of estrogen Receptor-α (ERα). At present, however, there Gonadotroph adenomas are the most common subtype of
are still no reliable commercial antibodies for Tpit. nonfunctioning adenomas (approximately three quarters in
Approximately, 30–40% of all surgically treated pitui- our series [6]). Although many of them are either negative
tary adenomas are clinically nonfunctioning adenomas or faintly positive for very few scattered cells or groups of
since they lack clinical and biochemical evidence of cells for gonadotropins, they can be reliably detected by
adenohypophyseal hormone excess. Morphologically nuclear immunoreactivity for SF-1 (Fig. 2). Most of them
they are heterogeneous and have been traditionally clas- are benign, slowly growing tumor presenting in middle-
sified into several histological subtypes by various meth- aged and elderly patients. The well-known common clini-
ods including immunohistochemistry using antibodies cal characteristics of nonfunctioning adenomas are those of
against adenohypophysial hormones, electron microscopy, this subtype.
in situ hybridization, or reverse hemolytic plaque assay. It
became evident that there are some differences in clinical Silent corticotroph adenomas
behavior and prognosis for each subtype. However, there
are limitations in achieving accurate classification using Traditionally, silent corticotroph adenomas had been dis-
hormone immunohistochemistry alone. An appropriate tinguished from other nonfunctioning adenomas by their
use of immunohistochemistry for pituitary transcription ACTH immunoreactivity. However, regardless of the func-
factors has a complementary role in obtaining an accu- tional status of the tumor, corticotroph adenomas, like
rate diagnosis for hormone-negative adenomas and, thus, nontumorous corticotrophs, express Tpit [4, 5]. In our pre-
can prevent them from being mistakenly classified as null vious study, one quarter of hormone-negative adenomas
cell adenomas [5, 6]. Subclassification of nonfunctioning were Tpit-positive with increased pro-opiomelanocortin
adenomas was revised accordingly. In clinical practice, an (POMC) mRNA expression [6]. Thus, ACTH-immuno-
accurate histological diagnosis should be obtained particu- histochemistry alone is insufficient to detect this subtype,
larly in nonfunctioning adenoma in young patients, with whereas assessment of Tpit expression is required for
frequent recurrence, and with aggressive clinical features the accurate diagnosis in some case (Fig. 3). This sub-
[6, 7]. type tends to show significant female preponderance and

13
Brain Tumor Pathology

Fig. 2  Nonfunctioning pituitary
adenoma in a 60 years old man.
The chromophobe adenoma was
immunonegative for adenohy-
pophysial hormones and pit-1,
but was SF-1-positive. The
diagnosis was a gonadotroph
adenoma

Fig. 3  Large and invading non-

functioning pituitary adenoma
in a 35 years old woman. The
chromophobe adenoma was
immunonegative for adenohy-
pophysial hormones, SF-1, and
pit-1, but was immunopositive
for Tpit. In RT-PCR, POMC
mRNA level was elevated (data
not shown). The diagnosis was a
silent corticotroph adenoma

are more frequently large macroadenomas with marked Silent adenomas of pit‑1 derivation
cavernous sinus invasion [6–8]. Surgical total resection is
often difficult and some of them are considered to behave This subtype consists of less than 10% of nonfunctioning
aggressively. adenomas and is in the family of somatomammothyrotrophs.

13
 Brain Tumor Pathology

They are more complex than other subtypes and are morpho- Prognostic clinical and pathological parameters
logical heterogeneous including plurihormonal pit-1-posi- and the definition of “aggressiveness”
tive adenomas (previously called subtype 3 adenomas [9]),
silent thyrotroph adenomas, silent somatotroph adenomas, Despite elimination of the term “atypical adenoma”,
et al. This subtype typically shows preponderance in younger assessment of markers of tumor proliferation (i.e. mitotic
patients. Of clinical importance, most of them are invasive, count and Ki-67 proliferation index) is recommended as an
large macroadenomas with a high Ki-67 proliferation index, important histological marker to predict recurrence, par-
and thus, tend to exhibit low resection rate and high recur- ticularly following incomplete adenoma resection [1, 2].
rence rate after surgery [7, 9]. However, no specific cutoff value is recommended. Clini-
cal parameters such as invasion assessed by MRI and/or
Null cell adenomas intraoperative examination should be considered an impor-
tant prognostic feature in identifying clinically aggressive
The definition of null cell adenomas was revised extensively adenomas [1, 2, 12].
in the new classification. This subtype is currently defined The definition of “aggressive” adenoma remains
as adenomas composed of adenohypophysial cells that do ambiguous, but it is generally assumed to represent some
not show any evidence of cell-type specific differentiation adenomas with rapid growth, significant invasion, large
using pituitary hormones, transcription factors, and ultras- and irregular configuration, and resistance to conventional
tructural features [10]. Hormone-immunonegative adenomas treatment, et al. In the 2017 WHO book, rapid growth,
account for 20–30% of nonfunctioning adenomas, whereas radiological invasion, and a high Ki-67 proliferation
null cell adenomas are quite rare, constituting less than 5% index are considered to represent clinical aggressiveness
of them (1.2% of nonfunctioning adenomas in our series and adenomas with these features are termed “high-risk”
[6]). Although the clinical data on this subtype, as it is cur- pituitary adenomas [3].
rently defined, are limited, they may represent an aggressive On the other hand, it has been pointed out that ade-
adenoma [6, 11]. noma subtypes are the most important determinant of
adenoma behavior [4, 5, 13]. Gonadotroph adenomas tend
to manifest particularly indolent behavior, especially in
Malignancy, aggressiveness, and prognostic older patients. In contrast, the 2017 WHO book indicates
parameters that there are some special adenoma subtypes that com-
monly show aggressive behavior. These include sparsely
Pituitary carcinoma (8272/3) granulated somatotroph adenoma, lactotroph adenoma in
men, Crooke cell adenoma, silent corticotroph adenoma,
Definition of pituitary carcinoma has not been changed in and plurihormonal pit-1-positive adenoma [1–3]. As has
the new WHO book. It is strictly defined as a tumor of ade- been mentioned in the former chapter, an accurate diag-
nohypophyseal cells that metastasizes craniospinally or is nosis of the latter two subtypes requires assessment of
associated with systemic metastasis. It must be emphasized the appropriate transcription factors, i.e., Tpit and pit-1,
that the definition is independent of the histological appear- respectively.
ance. No histological features can distinguish carcinoma
from ordinary adenoma prior to metastasis.
A new proposal: from pituitary adenoma
Elimination of the term “atypical adenoma” to “pituitary neuroendocrine tumor (PitNET)”

In the previous 3rd edition, pituitary adenoma was classi- Although most pituitary tumors behave benignly (i.e., do not
fied into typical adenoma (8272/0) and atypical adenoma metastasize) and are, therefore, called adenomas, they fre-
(8272/1). The latter was defined as adenomas with an ele- quently invade into the adjacent structures, leading to recur-
vated mitotic index and a Ki-67 proliferation index greater rence. One of the misleading in the current classification is
than 3%, as well as extensive nuclear immunostaining for a simplistic distinction between adenoma and carcinoma. In
p53. These adenomas with atypical morphologic features 2017, the International Pituitary Pathology Club proposed a
were considered to behave aggressively. However, the inci- new terminology, pituitary neuroendocrine tumor (PitNET),
dence of this category was relatively variable and prognos- which is consistent with that used for other neuroendocrine
tic significance could not be established despite more than neoplasms and which recognizes the highly variable impact
10 years of research on the utility of this classification [1, of these tumors on patients [14]. PitNETs are not simply
2]. Therefore, the term “atypical adenoma” was no longer endocrine diseases, but should be considered as tumors with
recommended in the new classification. endocrine manifestations within the context of oncology.

13
Brain Tumor Pathology

Conclusions 6. Nishioka H, Inoshita N, Mete O, Asa SL, Hayashi K, Takeshita A,

Fukuhara N, Yamaguchi-Okada M, Takeuchi Y, Yamada S (2015)
The complementary role of transcription factors in the accurate
The new WHO classification is very practical and mostly diagnosis of clinically nonfunctioning pituitary adenomas. Endocr
based on immunohistochemistry for pituitary hormones, Pathol 26:349–355
pituitary transcription factors, and other markers commonly 7. Nishioka H, Inoshita N, Sano T, Fukuhara N, Yamada S (2012)
Correlation between histological subtypes and MRI findings in
used in pathology practice [1]. In our previous study [6], the clinically nonfunctioning pituitary adenomas. Endocr Pathol
new classification demonstrated a good correlation between 23:151–156
the subtypes of nonfunctioning adenomas and clinical fea- 8. Scheithauer BW, Jaap AJ, Horvath E, Kovacs K, Lloyd RV, Meyer
tures. In the present article, the two major changes and a pro- FB, Laws ER Jr, Young WF Jr (2000) Clinically silent cortico-
troph tumors of the pituitary gland. Neurosurgery 47:723–730
posal of a new terminology, PitNET, were introduced. The 9. Erickson D, Scheithauer B, Atkinson J, Horvath E, Kovacs K,
new classification will hopefully benefit many pathologists, Lloyd RV, Young WF Jr (2009) Silent subtype 3 pituitary ade-
clinician, and patients harboring pituitary tumors. noma: a clinicopathologic analysis of the Mayo clinic experience.
Clin Endocrinol 71:92–99
10. Nishioka H, Kontogeorgos G, Lloyd RV, Lopes BS, Mete O, Nose
V (2017) Pituitary gland: null cell adenoma. In: Lloyd RV, Osa-
References mura RY, Klöppel G, Rosai J (eds) WHO classification of tumours
of endocrine organs, 4th edn. IARC, Lyon, pp 37–38
1. Lopes MBS (2017) The 2017 World Health Organization clas- 11. Balogun JA, Monsalves E, Juraschka K, Parvez K, Kucharczyk
sification of tumors of the pituitary gland: a summary. Acta Neu- W, Mete O, Gentili F, Zadeh G (2015) Null cell adenomas of the
ropathol 134:521–535 pituitary gland: an institutional review of their clinical imaging
2. Osamura RY, Lopes MBS, Grossman A, Kontogeorgos G, Trouil- and behavioral characteristics. Endocr Pathol 261:63–70
las J (2017) Introduction. In: Lloyd RV, Osamura RY, Klöppel G, 12. Trouillas J, Roy P, Sturm N et al (2013) A new prognostic clinico-
Rosai J (eds) WHO classification of tumours of endocrine organs, pathological classification of pituitary adenomas: a multicentric
4th edn. IARC, Lyon, p 13 case-control study of 410 patients with 8 years post-operative
3. Osamura RY, Lopes MBS, Grossman A, Matsuno A, Korbonits follow-up. Acta Neuropathol 126:123–135
M, Trouillas J, Kovacs K (2017) Pituitary adenoma. In: Lloyd 13. Kleinschmidt-DeMasters BK (2006) Subtyping does matter in
RV, Osamura RY, Klöppel G, Rosai J (eds) WHO classification pituitary adenomas. Acta Neuropathol 111:84–85
of tumours of endocrine organs, 4th edn. IARC, Lyon, pp 14–18 14. Asa SL, Casar-Borota O, Chanson P et al (2017) Pituitary ade-
4. Asa SL (2008) Practical pituitary pathology. What does the noma to pituitary neuroendocrine tumor (PitNET): an interna-
pathologists need to know? Arch Pathol Lab Med 132:1231–1240 tional pituitary pathology club proposal. Endocr Relat Cancer
5. Mete O, Asa SL (2012) Clinicopathological correlations in pitui- 24:C5–C8
tary adenomas. Brain Pathol 22:443–453

13