Pituitary TF–based classification improves diagnostic accuracy and prognostication of nonfunctioning PitNETs

Utility of pituitary transcription factors (TFs) improves the diagnostic accuracy and prognostication of nonfunctioning pituitary neuroendocrine tumours (NF-PitNETs), according to local data and a new diagnostic classification presented at the 5^th Annual Meeting of Endocrinology, Diabetes & Metabolism Hong Kong (EDM HK).

NF-PitNETs are PitNET subtypes without hormone hypersecretion. Despite absence or low levels of hormone secretion, NF-PitNETs express pituitary TFs such as steroidogenic factor-1 (SF-1), T-box pituitary transcription factor (TPIT) and pituitary transcription factor 1 (Pit-1), which can help disease classification. [J Clin Endocrinol Metab 2019;104:2473-2489]

In a study that included 113 patients with NF-PiTNETs treated in Queen Mary Hospital, Hong Kong, the most common lineage was SF-1 (58 percent), followed by TPIT (19 percent), while 16 percent of patients were TF-negative with no distinct lineage (ie, null cell adenoma or plurihormonal tumour). “Notably, the concordance rate of [immunohistochemistry (IHC) staining between] hormone and TF was only 23.9 percent, which is quite poor,” reported Dr Chariene Woo of Department of Medicine, University of Hong Kong.

Consistent with Hong Kong’s findings, a Japanese study with 516 NF-PitNET patients revealed SF-1 and TPIT in 66.4 percent and 26.9 percent of the hormone-negative group (n=119), respectively. Only 5 percent were TF-negative without distinct lineage. [Endocr Pathol 2015;26:349-355] “These results indicate that IHC staining for pituitary TFs may increase the diagnostic accuracy in NF-PiTNETs,” Woo commented.

Dichromatic staining for nuclear pituitary TFs may contribute to improved PiTNET diagnosis. “[For example,] in gonadotrophic tumour, SF-1 shows a distinct staining pattern, but cytoplasmic follicle-stimulating hormone [FSH] reveals a scarce and scanty pattern,” Woo explained. [Endocr Pathol 2022;33:6-26] “If only IHC staining for FSH is performed, NF-PiTNETs may be misclassified as null cell tumour [ie, PiTNETs expressing neither hormone nor TF].”

Of note, NF-PiTNETs with TPIT lineage and null cell adenoma are recognized as aggressive and invasive PiTNET subtypes. In contrast, the SF-1 lineage may be associated with favourable outcomes, such as higher rates of gross total resection, as well as less invasive and recurrent disease. [Acta Neurochir (Wien) 2021;163:3143-3154; Endocr Pathol 2015;26:63-70; Neurosurg Focus 2020;48:E13]

“Our analysis also revealed increased rates of invasive tumours in NF-PiTNETs with TPIT lineage [odds ratios (OR), 4.3; p=0.04] vs other distinct lineages. SF-1 lineage was associated with higher complete resection rate [OR, 2.6; p=0.037], with better disease-free survival,” Woo reported.

“In NF-PiTNETs without distinct lineage, less complete resection was achieved [OR, 0.3; p=0.036] and higher rates of recurrent or residual disease were found [hazard ratio; 3.02; p=0.002] vs tumours with distinct lineages,” she continued. “With accurate classification of PiTNETs, particularly after complete resection, prognostication and clinical management can be further improved.”

For accurate and cost-effective PiTNET diagnosis, an Australian team has proposed an algorithm that omits hormone IHC in SF-1–positive or TPIT-positive samples. However, hormone IHC is still required in those with Pit-1 positivity. In the study cohort (n=113), this algorithm demonstrated 100 percent concordance with the gold standard (ie, a full panel of hormone and TF IHC), with the lowest average number of immunostains and lowest average cost vs other algorithms. [Pituitary 2022;25:997-1003]

“While establishment of diagnostic algorithm is ongoing, this study can be a potential diagnostic model for PiTNETs,” Woo commented.