Introduction

The most commonly occurring primary mesenchymal tumor in the gastrointestinal tract is a gastrointestinal stromal tumor (GIST). Since the tumor has been molecularly characterized only recently, it is not easy to determine its actual frequency; however, population-based studies suggest that the annual incidence per million population is between 11 and 151. Most GISTs seemingly occur sporadically, but nearly 5% of all GISTs are linked with inheritable mutations or syndromes.

Several studies reported the association between GIST and second primary malignancies. Next to other gastrointestinal tumors, genitourinary (GU) cancer is associated with GIST2,3,4,5,6. In addition, the association is often related to KIT mutations7,8.

The most reported associated primary GU tumors are prostatic adenocarcinoma, renal cell carcinoma (RCC), urothelial cancer, seminoma, and adrenal tumors. Therefore, we set out to identify the prevalence of genitourinary tumors in our hospital patients with histopathological diagnosis of GIST.

Material and methods

This retrospective study included all patients diagnosed with GIST in the pathology department database. The association with RCC and other genitourinary tumors was identified. We defined genitourinary cancer as any cancer involving the kidney, adrenals, urinary bladder, prostate, testis, or penis. We cross-examined the RCC database for the association of GIST. Inclusion criteria included any patient with a histopathologic diagnosis of primary GIST. We excluded duplicate pathology reports of the same patient and cases with GIST recurrences. We excluded tumors of the ovaries and the female genital tract. Descriptive statistics were used. A systematic review literature search was made in Medline/PubMed. The search syntax was generated to combine MeSH terms of “GIST” OR “Gastrointestinal Stromal Tumors” AND MeSH terms of “Neoplasms, Second Primary” OR “Neoplasms, Multiple Primary” OR a text word indicating genitourinary cancer (renal, adrenal, urothelial, bladder, prostatic, OR testicular cancer) OR a genitourinary organ (kidney, adrenal, prostate, urinary bladder, ureter, OR testis). A filter was added restricting the search to the English language and humans.

Ethical approval

This retrospective chart review study involving human participants was following the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments. The Human Investigation Committee (IRB) and Research Ethics Committee of King Faisal Specialist Hospital and Research Center approved this study.

Consent for publication

The Human Investigation Committee (IRB) and Research Ethics Committee of King Faisal Specialist Hospital and Research Center waived the consent because of the study's retrospective nature, and all the procedures being performed were part of the routine care.

Results

Our patients: From 2003 to 2020, 170 patients had a histopathologic diagnosis of primary GIST, 100 men and 70 women, median age of 57 (range 9–91) years at the time of diagnosis. There were 144 patients with GIST-only tumors (84.7%), 26 with other primary cancer, including six patients with primary genitourinary cancer and two with GIST direct invasion of the urinary bladder (Table 1). Primary GIST was most commonly gastric, spindle cell type, had a low mitotic rate and CD117, DOG1, or both immunostainings. Few cases underwent genetic mutation analysis, and none had genetic counseling. Only one-third of patients were treated with imatinib. Patients with another primary cancer had significantly lower TNM stage and less epithelioid histopathology than GIST-only patients. No significant differences were observed amongst patients with GU vs. no GU cancer.

Table 1 Demographic and pathological characteristics of patients with GIST and associated cancer.
Full size table

All GIST – GU tumors in our cohort were synchronously discovered during the investigation and treatment of the GU cancer. Two patients had associated clear cell RCC. One tumor was a renal mass with radiological characteristics suggesting papillary RCC (Fig. 1). The tumor was 8 mm in diameter. As there was no indication for surgery, it was followed up by observation with no histological diagnosis. Follow-up CT after two years was consistent with RCC (Fig. 2). Cross-examination of the RCC database retrieved the two histologically proven RCCs. Three cases had adrenal tumors (one adrenal carcinoma, one an isolated adrenal GIST, and one pheochromocytoma). One tumor presented as a urinary bladder mass and was evaluated by transurethral resection. The bladder tumor on histopathology was GIST due to direct invasion from the colon that was not clear on the preoperative radiological diagnosis. Another patient had a urinary bladder invasion by GIST arising from the mesentery.

Figure 1
figure 1

Radiological diagnosis of RCC in a GIST patient. MRI showing a coronal section of the abdomen with renal lesion (arrows). (a) T1 FS image showing a mild low signal intensity of a right posterior 8 mm renal mass. (b) T2 FS image showing increased intensity of the renal lesion. (c) T1 FS image showing hyperenhancement of the lesion after contrast injection. d. ADC MAP showing diffuse restriction of the lesion.

Full size image
Figure 2
figure 2

CT scan of the abdomen at follow up of the same patient after 2 years showing Hounsfield density measurement of the renal tumor consistent with right RCC (arrows). (a) CT without contrast, lesion density 33 ± 13. (b) Arterial phase, the density increased to 138 ± 12. (c) Venous phase the density decreased to 118 ± 13. (d) Delayed phase the density decreased to 94 ± 16.

Full size image

Of note is that the GIST cohort included 63 male patients who were 60 years or older (median 71 ± 8.7 years, range 60–94). None of them, however, had prostate cancer.

Systematic review: A total of 429 articles were identified, and abstracts were reviewed. Further selection was based on information indicating the presence of GIST in association with genitourinary cancer or organ. Duplicate publications were removed. We found 59 articles meeting the selection criteria and added ten more from references in these articles. A total of 69 articles were reviewed. Seventeen articles reported the prevalence of genitourinary cancer associated with GIST ranging 2.1–15.6% (Table 2)7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23. Prostatic adenocarcinoma was the most common primary genitourinary cancer (2.1–14%), followed by RCC (0.6–11.4%). Rarely seminoma8,11,16,21 and a single case of adrenal neuroblastoma were reported19. Additional 52 articles included case reports of the association (Table 3)2,3,5,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55, or a primary genitourinary organ GIST (Table 4)56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72.

Table 2 Reported series of GU cancer associated with GIST.
Full size table
Table 3 Case reports of GU cancer associated with GIST.
Full size table
Table 4 Case reports of primary GU organ EGIST.
Full size table

There were 35 case reports of an association between GIST and genitourinary cancer. The most common associations were prostatic cancer (12), adrenal pheochromocytoma (11), and RCC (10). Case reports of primary extra gastrointestinal stromal tumor (EGIST) affecting the genitourinary tract were 17. The most common genitourinary EGIST was affecting the prostate (n = 15). The radiological and pathological evidence suggested that the GIST was not an extension from the adjacent bowel. Two cases were reported from the adrenal; one was from our institution56,57.

Discussion

The number of our patients ranks 9th among the 17 studies reporting the prevalence of GU cancer in GIST patients (Table 2). In contrast, we report a 3.5% association ranking our study among the four least reported associations. The lower rate in our patients is probably related to the absence of prostate cancer compared to others. Series that reported prostate adenocarcinoma had a mean and median age between 65 and 777,9,13,15,16. In our study, 61.3% of men were 60 years or older (Table 5). Those men had a median age of 71 and ranged from 60 to 94. At this age range, we expected to identify some cases of prostate cancer, but we did not find any73. The discrepancy is probably related to the high incidence and prevalence of prostate cancer in USA cancer registries compared to the Saudi and worldwide population (Table 6)74. There is a marked discrepancy in the cancer prevalence of the genitourinary tract74. The prevalence rate per 100,000 population of cancers in the USA is threefold higher than the reported worldwide rate and seven times the Saudi Arabian rate (Table 6). Several factors, like geographical differences in genetic makeup, risk factors, early detection, reporting period, and reporting system, may explain these differences. Of note is that the reported prevalence of prostatic cancer is much higher in patient populations than in the community. For example, in the USA, the prevalence of prostate cancer in the community for men 60 years and above is 1.96%, whereas the corresponding prevalence in the Surveillance, Epidemiology and End Results (SEER) database is 4.8–16.7%73,74. As our report is in a patient cohort rather than a community-based study, we believe that the lack of prostate cancer detection is an actual difference compared to other western countries. These findings may indicate that the association between prostate cancer and GIST is not due to a common risk factor but rather a geographical difference for prostate cancer. Another discrepancy is the absence of primary urothelial cancer in our cohort. Others reported 0.9–2.9% prevalence associated with urothelial cancer8,13,15,16,17,20. This feature is also surprising as urothelial cancer is highly prevalent in Saudi Arabia, comparable to renal cancer (Table 6).

Table 5 Men 60-year-old and above in the current study and corresponding USA prostate cancer prevalence.
Full size table
Table 6 Estimated number of 5-year prevalent cases in 2020 of genitourinary tumors according to the geographical region, both sexes, all ages74.
Full size table

On the other hand, the prevalence of 1.8% of renal cancer is comparable to most studies (0.6–2.2%), except two reporting 4.35% and 11.4%8,12. Out of the three cases, two were CRCC, and the radiological features of the third were suggestive of papillary RCC. This is not surprising as papillary RCC prevalence in GIST patients is much proportionally higher14,23. The GISTs in our GU cancer patients were incidental during planned nephrectomy or adrenalectomy. The incidental GIST was small but otherwise pathologically comparable to GIST with no associated GU cancer (Table 1). No genetic analysis was available to link the two pathologies. Clinically, however, none of our patients had VHL or NF-1 disease.

Other unique features of this study involve a rarely reported association with adrenal tumors. Of interest is the primary GIST of the adrenal gland. There is only one reported adrenal EGIST other than the one reported from our institution (Table 4)56,57. There is no radiological evidence that the adrenal GIST was a direct or distant spread of another primary GIST lesion. We also report two adrenal tumors, one carcinoma, and the other pheochromocytoma, none reported in the reviewed prevalence series (Table 2). However, sporadic cases of 11 pheochromocytomas were reported and most associated with NF1, unlike our patient (Table 3). There is only one adrenal carcinoma case report in addition to the one in our series (Table 3)35. Rarely was urothelial cancer reported in association with GIST (Table 2). Although one case presented as a bladder tumor, unlike the other reports, the two bladder tumors we report were secondary to direct invasion of a primary intestinal GIST. In addition, unlike the reports, we did not find associated testicular tumors. The strong point of this study is that it includes a large cohort of patients with histopathology-proven GIST comparable to large international series. In addition, it highlights prevalence differences unique to our geographical location. The study's weakness is that it did not include molecular and genetic profiling of the tumors and its retrospective design.

Conclusion

We report the rare association between GIST tumors and primary genitourinary cancer, mainly RCC and adrenal tumors. Also, we identified a secondary invasion of the urinary bladder. Unlike the reported series, none of the older male patients had clinical prostate cancer.