COW-2021-45

CASE OF THE WEEK

Editors: Francesca Khani (Frk9007@med.cornell.edu) and Mahmut Akgul (akgulm@amc.edu)

2021-45/November 8
Contributors: Laurence A Galea, Daniel Moon, Darren J Katz

A man in his 60s presented with lower urinary tract symptoms. After imaging studies and multiple transurethral resections of the prostate, he ultimately underwent robot-assisted radical prostatectomy.

Quiz

1. What is the correct diagnosis?

a. Leiomyosarcoma

b. Primary prostatic gastrointestinal stromal tumour (GIST)

c. Solitary fibrous tumor

d. Schwannoma

e. Prostatic stromal tumour of uncertain malignant potential (STUMP)

Prostatic stromal tumour of uncertain malignant potential (STUMP)

The patient initially underwent transurethral resection of the prostate (TURP) for lower urinary tract symptoms. Histological examination of the TURP chips showed prostatic adenocarcinoma, Grade Group 1 (Gleason score 3+3=6) involving 10% of the tissue and benign myoadenomatous hyperplasia. A follow up prostate MRI showed a PI-RADS 5, 7cm complex solid and cystic mass near completely replacing the entire prostate, with features raising the possibility of atypical histology, including sarcoma (Fig 1A). A deeper diagnostic TURP was carried out, of which histology showed features in keeping with prostatic stromal tumour of uncertain malignant potential (STUMP), with a comment stating that STUMP can occur in association with prostatic stromal sarcoma (PSS)(not shown).

The patient then underwent robot-assisted radical prostatectomy. Macroscopically, the tumour replaced most of prostate gland, 75mm in maximum dimension, with some residual transitional zone and anterior fibromuscular zone. It had a circumscribed, unencapsulated multinodular yellow gelatinous and white-tan, fleshy cut surface (Fig 1B). Histologically, the tumour was mildly to moderately cellular and composed of a haphazard arrangement of bland spindle cells in mostly myxoid and focally collagenous stroma. Focally, the cells showed enlarged, pleomorphic, occasionally multinucleated, degenerative, smudgy hyperchromatic nuclei (Figs 2-4). Mitotic figures were sparse with less than 5 mitoses identified throughout the whole tumour. No atypical mitotic figures were seen. There were no intermingled prostatic glands. There were no features to suggest co-existent PSS (i.e. no prominent mitotic activity, no tumour necrosis apart from that attributed to previous TURPs, and no overt stromal overgrowth). The lesional cells were positive for CD34, H-caldesmon, and androgen receptor. They were negative for cytokeratins(AE1/AE3and CKMNF 116), c-kit, SMA, desmin, myogenin, S100, SOX10, STAT-6, ALK-1, MUC-4 and PR. The Ki-67 index was less than 5%. A diagnosis of prostatic STUMP was rendered.

STUMP is a tumor of the specialised prostatic stroma. There are five histological pattern of STUMP: (1) hypercellular stroma with scattered atypical degenerative-looking cells intermingled with benign glands; (2) hypercellular stroma without atypical cells intermingled with benign glands; (3) hypercellular stroma without atypical cells that is not intermingled with benign glands; (4) phyllodes-like pattern with variably cellular fibrous stroma, with or without atypia covered by benign-looking prostatic epithelium; and (5) extensively myxoid stroma that lacks the nodularity and thick walled blood vessels of benign nodular stromal hyperplasia.

There are no specific immunohistochemical markers for STUMP. They are positive CD34. Progesterone receptor is frequently expressed, although estrogen receptor expression is uncommon. The lack of smooth muscle markers (SMA and desmin), c-kit, STAT-6 and S100 exclude leiomyosarcoma, primary prostatic gastrointestinal stromal tumour, solitary fibrous tumor and schwannoma respectively.

There are no validated morphological criteria to distinguish between STUMP and PSS, though prominent mitotic activity, tumor necrosis and overt stromal overgrowth are worrying features, and the presence of two or more of these features is highly suggestive of PSS. STUMP and PSS can co-exist in the same specimen. STUMPs can also recur as PSS or progress to PSS during follow-up. In a recent article, Acosta AM at al. showed that STUMP and PSS are molecularly heterogeneous and include neoplasms that harbour genetic aberrations seen in specific mesenchymal tumours arising nother anatomic sites, including soft tissue and the uterus. They suggested that such tumours may not represent a single disease entity or a specific disease group arising from the prostatic stroma.

1. Mooch H, Humphrey PA, UlbrichtTM, Reuter VE. WHO Classification of Tumours of the Urinary System and Male Genital Organs.4th ed. Lyon, France: IARC Press; 2016.

2 Herawi M, Epstein JI. Specialized stromal tumors of the prostate: a clinicopathologic study of 50 cases. Am J Surg Pathol.2006 Jun;30(6):694-704.

3. Acosta AM, Sholl LM, Dickson BC, McKenney JK, Gordetsky JB, Pins MR, Marino-Enriquez A, Dong F, Dubuc AM, Cin PD, Fletcher CDM. Re-evaluating tumors of purported specialized prostatic stromal origin reveals molecular heterogeneity, including non-recurring gene fusions characteristic of uterine and soft tissue sarcoma subtypes. Mod Pathol.2021 Sep;34(9):1763-1779.

Laurence A Galea
Melbourne Pathology (Sonic Healthcare)
Victoria, Australia
Email: Laurence.Galea@mps.com.au
Twitter handle: @DrLaurenceGalea

Daniel Moon
Australian Urology Associates, Epworth Healthcare, and University of Melbourne
Victoria, Australia

Darren J Katz
Men’s Health Melborne, University of Melbourne, and Western Health
Victoria, Australia

Prostate

Prostatic stromal tumour of uncertain malignant potential, STUMP, prostatic stromal sarcoma.