International case series of metastasis to penis

Penile metastases are rare, with about 400 reported via case reports or small series. Under such circumstances coherent data about their behaviour may not be possible given vagaries of publication. For instance, one published meta‐analysis has proposed that malignant priapism contributes to worse survival, but this may be due to selection bias in published reports. We sought to evaluate clinical characteristics associated with survival in an international cohort of patients with metastases to the penis treated in major genitourinary cancer programmes.

and perineal urethrostomy 3 for penile cancer.We sought to leverage their extensive experience in the management of secondary metastatic disease of the penis.

| PATIENTS AND METHODS
After approval by the appropriate Institutional Review Boards, records of the collaborating centres were screened for men presenting with metastatic disease to penis.Lesions that resulted from direct extension from any cancer were excluded.Parameters of specific interest included primary site, synchronous versus metachronous presentation, other sites of metastasis, metastatic site within penis, existence of priapism, treatment of penile metastasis, local control duration and survival after penile metastasis.The Kaplan-Meier method was used to generate an overall survival curve.A multivariable Cox proportional hazard model was used to calculate hazard ratios (HRs) based on a panel of covariates determined a priori as mentioned above.All tests were two-sided, and p < 0.05 was considered statistically significant.
All analyses were performed using SPSS software system 2021, Version 28.0 (IBM, Armonk, NY).
As might be expected given the rarity of the occurrence, no clinical parameter including primary histology, the presence of synchronous or metachronous metastases, additional metastatic sites or priapism showed statistical survival benefit or detriment.

| DISCUSSION
The mechanism of malignant disease spread to penis has been proposed as venous or lymphatic spread, 4 based on the robust plexus involving pelvic organs and the dorsal penis.As early as 1919, Ewing noted: 'The mechanism of the circulation will doubtless explain many of these peculiarities, for there is as yet no evidence that any one … organ is more adapted than others'. 5Batson, in 1942, described an injection study in cadavers that yielded 'valveless vessels which carr [y] blood under low pressures, … constantly subject to arrests and reversals in the direction of the flow of blood'.This phenomenon was termed 'retrograde transport'. 6While this may be true in some cases, more recent data suggest this likely represents too simple a mechanism.Work by Fidler et al. subsequently described site specificity of metastasis of different cell lines, 7 leading most to reconsider the words of Paget in 1889, describing non-random patterns of metastasis: 'When a plant goes to seed, its seeds are carried in all directions; but they can only live and grow if they fall on congenial soil'. 8 the other hand, local interruption of lymphovascular flow could certainly be responsible for priapism associated with penile metastasis.Early reports, including one by Young 9 described priapism related to primary penile cancer.Malignant priapism due to metastasis was first described in 1928 by Begg. 10 Lin et al. provided a modern update of the phenomenon in 2011. 4The frequency of priapism as a presenting symptom in that review was quite frequent (20%-50%), although this may be due to selection bias in published reports.In the current work, overall frequency of priapism was 21% (n = 7).Both Lin et al. 4 and Cocci et al. 1 proposed that malignant priapism contributed to 'grim' prognosis.Our data certainly concur: Median survival of such patients was 6.2 months (range 0.9-15.4m), although not significantly different from patients without priapism (Mann-Whitney p = 0.11).
Not unexpectedly, therapy for such lesions are a function of patient performance status and disease elsewhere.While penectomy is an attractive option clinically for the local disease, it is far less prudent in the presence of multiple other metastases or for an asymptomatic patient otherwise doing poorly.Such patient selection contributed to local control in all but one of our total penectomy cases, but a corresponding survival benefit was not observed (Mann-Whitney p = 0.16).
Prior to this report, almost 400 cases of metastasis to penis have been reported since 1870. 4Even with the centres of excellence represented here, problems arose both with recalling and with collecting cases since penile metastases are unlikely to be a specific parameter in any database, and resources may not exist to manually retrieve them.Using an automated data retrieval system for the ICD9 code of 198.82 ('Secondary malignant neoplasm of genital organs') includes 15 categories for both genders.Finally, even if further sub-coded by male gender, results include the enormous population of patients with metastatic prostate cancer as well.Commentary made by the contributors included many urethral primary lesions and of primary penile melanoma (n > 20) or sarcoma (n = 2) cofounding search results.
A specific clinical phenomenon not included here were two cases of extramammary Paget's disease following treatment for urothelial carcinoma of the bladder.We did not consider these strictly to be metastases, since these cases are likely due to intraepithelial spread, not hematogenous.Nineteen other similar cases have been reported separately. 11

| CONCLUSION
Metastasis to penis arises most frequently from pelvic primaries, although there may be a data retrieval bias in these centres for genitourinary primaries rather than other lesions such as lymphoma.
Priapism does not appear to correlate with survival in this large, welldefined series.
Presenting parameters of collected patients.