Why is pituitary carcinoma so rare?

KEYWORDS:

• Pituitary adenoma
• pituitary carcinoma
• aggressive pituitary tumor
• cancer
• oncogene-induced senescence

Pituitary adenomas (PA) are rather frequent, they constitute up to 25% of all intracranial tumors, with a prevalence varying from 70 to 115 cases per 100,000 persons and an incidence of 4 to 6 per 100,000 persons per year [1]. The prevalence among autopsy and radiological studies varies between 10% and 20%. Pituitary carcinomas (PC), on the other hand, are among the least frequent malignancies known to man, barely reaching 0.2–0.4% of all pituitary tumors [2]. In fact, most practicing endocrinologists worldwide do not get to see a single case of PC during their professional lifetime. Published information regarding PC mostly consists of isolated case reports, very few series, and a handful of reviews [3–5]. Over 70% of PC are functioning, usually ACTH- or PRL-secreting tumors that arise from apparently benign, but usually invasive macroadenomas, with a latency period from adenoma to carcinoma that varies from 6 months to over 10 years [3–5]. Features associated with malignancy such as hypercellularity, nuclear pleomorphism, increased mitotic activity, dural, bone and vascular invasion, extracellular matrix degradation, neoangiogenesis, a high Ki-67 proliferative index, and p53 immunostaining are not always present in PC and can occasionally be found in nonmalignant pituitary tumors [4]. Thus, the documentation of craniospinal or distant metastasis (bone, lymph nodes, liver, and lung, being the most frequently reported sites) is a sine qua non criteria to establish the diagnosis [2–5].

Lying between PA and PC in the pathological spectrum of pituitary neoplasms, we found what used to be called ‘atypical pituitary adenomas’ and that are currently known as aggressive pituitary tumors (APT). According to the European Society of Endocrinology (ESE) 2018 guidelines, APT are defined as tumors with radiological evidence of invasiveness, particularly cavernous sinus invasion, with an unusually rapid growth rate and that is persistent despite multimodal therapy (surgery, radiation, pharmacological treatment) [6,7]. It is important to emphasize, however, that while an aggressive tumor is almost always invasive, an invasive lesion does not always behave aggressively. The latter is illustrated by some giant prolactinomas that may be quite sensitive to pharmacological treatment with dopamine agonists. Interestingly, the conversion of a nonfunctioning into a functioning tumor, or more specifically, of a silent corticotrope adenoma into a Cushing disease-causing tumor is associated with an aggressive biological behavior and should warn us of the possibility of an eventual malignant transformation [8].

In an attempt to summarize the complex cellular and subcellular events that underlie malignant transformation, Hanahan and Weinberg published back in 2000 their seminal article ‘The Hallmarks of Cancer’ [9]. The original review is based on six biological processes that characterize carcinogenesis, namely, 1) self-sufficiency in growth signals, 2) insensitivity to anti-growth signals, 3) apoptosis evasion, 4) limitless replicative potential, 5) sustained angiogenesis, and 6) tissue invasion and metastasis [9]. An update published a decade later by the same prestigious scientists added four other features or traits: a) reprogramming energy metabolism, b) evading immune response, c) genomic instability and mutation, and d) tumor-promoting inflammation [10]. In 2022, Hanahan incorporated still other hallmarks related to phenotypic plasticity and disrupted differentiation, tumoral microenvironment and senescence, non-mutational epigenetic reprogramming, and polymorphic microbiomes [11]. Although Hanahan and Weinberg views have remained an essential classic of the cancer biology literature, they have not been free of controversy. Part of the criticisms received by the Hallmarks of Cancer has to do with the fact that, except for the ability to give rise to metastasis, benign and malignant neoplasms share most of the biological features that these authors defined as cancer hallmarks [12]. That benign and malignant neoplasms, at least initially, share similar biological processes of oncogenesis opens a rather interesting paradigm for pituitary tumors, which are known to be quite resistant to malignant transformation.

Over 50% of PC are functioning ACTH-secreting tumors [2–5]. We recently had the opportunity to evaluate through whole exome sequencing 10 pituitary tumors, representing the whole pathological spectrum of the corticotrope and including four Cushing disease-causing corticotrophinomas, three silent corticotrope adenomas, one ACTH-secreting macroadenoma in a patient with Nelson’s syndrome, one Crooke cell macroadenoma, and one ACTH-secreting carcinoma that evolved from a microadenoma over the span of 15 years [13]. We found, previously reported, recurrent somatic mutations in six pathogenically relevant genes: HSD3B1 (all 10 tumors), TP53 (all except one silent ACTH adenoma), CDKN1A (6 tumors), EGFR (6 tumors), AURKA (4 tumors), and USP8 (2 tumors) [12]. The PC was the tumor with the highest number of genomic abnormalities, including single nucleotide polymorphisms and copy number variations [13]. Interestingly, we carried out a phylogenetic inference analysis trying to establish a hypothetical sequential step transformation from a silent corticotrope adenoma to a functional tumor and finally, to an ACTH-secreting carcinoma [13]. We found two distinct clades. The first clade, characterized by expressing a SNV of the gene encoding the activating transcription factor 7-interacting protein (rs.3213764), includes 2 of the 3 silent adenomas and 2 of the 5 corticotrophinomas causing Cushing disease [12]. These four tumors have the same SNV profile; therefore, we can assume that they harbor all the genes that must be altered to allow the transition from a silent adenoma to a clinically eloquent tumor [13]. The second and largest clade comprises the Crooke cell adenoma, two ACTH-adenomas causing Cushing disease, one of the silent ACTH-adenomas and the ACTH-secreting carcinoma [13]. All the tumors in this clade harbor the necessary molecular alterations to evolve from a functioning ACTH adenoma to a malignant corticotrophinoma and are characterized by expressing a polymorphism of the gene encoding the DNA mismatch repair protein (rs.1650697) [13].

Indeed, several hypotheses have been advanced to explain PA resistance to malignant transformation, although none of them is fully satisfactory [14]: 1) PA seem to have a notoriously high apoptotic rate, yet this cannot explain by itself why the adenoma maintains a stable size over the years; 2) PA are poorly vascularized lesions, therefore a limited angiogenesis is likely to partially contribute; 3) Pituitary cells are irreversibly committed to terminal differentiation, which in theory by itself makes malignant transformation more difficult to occur; yet, tumor stem cells have been found in PA and appear to have a role in tumor initiation, progression, and invasion as well as in the resistance to pharmacological agents such as somatostatin analogs [15]; 4) The tumor microenvironment may modify the course of a pituitary tumor through the elimination of abnormal clones by cytotoxic T lymphocytes and Natural Killer cells; and 6) A process called oncogene-induced senescence, whereby the same driver oncogenic mutations that are supposed to confer a proliferative stimulus, for some reason at some point, trigger the activation of different cell cycle regulators such as p16 (CDKN2A), p53 and pRB (retinoblastoma protein) that repress E2F (steroidogenic factor 2) and result in cell cycle arrest. When this happens, cells acquire a ‘senescent’ phenotype consisting of the cytoplasmic accumulation of lysosomal acid-β-galactosidase and heterochromatin foci [16,17]. In this context, GH-secreting adenomas are known to express the highest levels of nuclear p21 and p16 as well as cytoplasmic acid-β-galactosidase immunostaining [16,17].

In conclusion, the extreme rarity of pituitary carcinomas is likely the result of multiple, complex, and highly efficient mechanisms that prevent malignant transformation through the promotion of cell proliferation arrest in the context of oncogene-induced senescence. Cell proliferation control by immune cells and their different cytokines present in the tumoral microenvironment, as well as an inherent limitation of angiogenesis are likely to play important roles in this process. A better understanding of these mechanisms is likely to generate crucial knowledge that transcends pituitary oncogenesis, potentially resulting in the identification of molecular targets for the treatment of other malignant neoplasms.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgment

(anonymized) participates in the Program ‘Investigadores e Investigadoras por México’ of National Council of Science and Technology.

Additional information

Funding

The authors have no funding to report.