We have read with great interest the comments made by Benesch and Urban on our recent description of the distinctive features of primitive neuroectodermal tumor (PNET) of the kidney.1 We agree that PNET of the kidney should be considered part of the Ewing's sarcoma family of tumors (ESFT), sharing the same histologic, cytochemical, cytogenetic, and molecular features.2, 3 Our purpose in describing these patients was twofold: first, we wished to inform oncologists and pathologists that PNET may present as a primary renal tumor; and second, we suggested that PNET arising in the kidney may have an aggressive clinical course. PNET arising primarily in the kidney is rare. As our series illustrates, this entity must be included in the differential diagnosis of primary renal tumors, particularly in children and young adults. Furman et al.4 noted that lymphoma, carcinoid tumors, small cell neuroendocrine carcinoma, and rhabdomyosarcoma may present as primary small cell tumors of the kidney. As we discussed, Wilms' tumor of blastemic type, renal neuroblastoma, and malignant rhabdoid tumor may share histologic features with PNET. Benesch and Urban reiterate our recommendation that immunohistochemistry for CD99 is mandatory in such cases, and we agree that molecular characterization of these tumors should be performed when tissue is available. Four other recent case reports also note the rarity of this tumor and the importance of considering renal PNET a distinct clinical entity.4-7 Of the 11 cases described (8 with metastatic disease at diagnosis), 1 patient died after surgery, 7 patients died of disease progression at a median of 5 months after diagnosis, and the 3 survivors were still receiving treatment at the time the reports were made.1, 4-10 Given the small number of cases of PNET of the kidney, statistically significant comparisons of the outcomes of patients with ESFT by this primary site are not possible. Nevertheless, PNET of the kidney appears to represent a subgroup with very aggressive behavior. We agree that patients with advanced local or metastatic disease are at higher risk of overall treatment failure; however, PNET, arising at other sites typically respond well to initial therapeutic interventions. Kushner et al.11 reported a 20-year retrospective series that excluded a number of long term survivors with PNET "because their small tumors were inadequate for confirming the diagnosis." In that report, 9 of 11 patients who received more that 1200 mg/m2/course of cyclophosphamide initially responded to chemotherapy. Using very aggressive short term therapy, this group recently reported initial responses in all 32 evaluable patients with peripheral PNET, 77% event free survival in 24 patients with locoregional disease, and 5 of 6 patients with lung metastases progression free.12 This report suggests that patients with metastatic peripheral PNET may be curable with aggressive therapy. In other reports, 13 of 14 patients13 and 16 of 17 patients14 with peripheral PNET responded to initial combination chemotherapy. Although we previously reported poor outcome for patients with PNET of soft tissues,15 we have recently updated these data.16 In our recent group of 17 patients with soft tissue PNET treated since 1988, all 9 patients evaluable for initial response to chemotherapy responded. The 5-year progression free survival is 62 ± 16% for the entire group, including 8 patients with tumors >8 cm and 3 with metastatic disease at diagnosis. In contrast, 2 of our patients with renal PNET, all diagnosed after 1990 and receiving modern, aggressive therapy, showed primary resistance to agents that are very active against the ESFT, dying 1 and 4 months after diagnosis. The patients described by Scheaff et al.,6 Chan and Llewellyn,8 and Mor et al.,9 2 with localized disease and 1 with disease spread only to retroperitoneal lymph nodes, all died within 10 months of diagnosis. Our series and the four other new case reports4-7 suggest that PNET of the kidney is a distinct clinical entity that must be distinguished from other primary renal tumors. Therapeutic approaches for this tumor should be the same as for other sites of ESFT, but may be less successful. Carlos Rodriguez-Galindo M.D.*, Neyssa M. Marina M.D. , David M. Parham M.D. , William H. Meyer M.D.?
51