Renal Failure in the Denys-Drash and Wilms' Tumor-Aniridia Syndromes1

[CANCER RESEARCH 60, 4030–4032, August 1, 2000] Advances in Brief

Renal Failure in the Denys-Drash and Wilms’ Tumor-Aniridia Syndromes1

Norman E. Breslow,2 Janice R. Takashima, Michael L. Ritchey, Louise C. Strong, and Daniel M. Green Department of Biostatistics, University of Washington, Seattle Washington 98195-7232 [N. E. B.]; Fred Hutchinson Cancer Research Center, Seattle, Washington 98109 [J. R. T.]; Department of Surgery, University of Texas [M. L. R.] and Department of Experimental Pediatrics/Genetics, M. D. Anderson Cancer Center [L. C. S.], Houston, Texas 77030; and Department of Pediatrics, Roswell Park Cancer Institute, Buffalo, New York 14203 [D. M. G.]

Abstract Patients and Methods

Nearly 6000 patients enrolled in four clinical trials of the National Wilms’ Patients. The study population consisted of 5976 patients enrolled between Tumor Study Group during 1969–1995 were followed until death or for a 1969 and 1995 at age 15 years or under in one of the first four NWTSG protocol median of 11.0 years of survival for the onset of renal failure (RF). Thirteen studies. All patients had a diagnosis of Wilms’ tumor of favorable (5572) or of 22 patients with Denys-Drash syndrome and 10 of 46 patients with the anaplastic (404) histology according to the NWTSG Pathology Center (10). Wilms’ tumor aniridia syndrome developed RF. The cumulative risks of RF Congenital anomalies and syndromes including WAGR and DDS were ascer- at 20 years from Wilms’ tumor diagnosis were 62% and 38%, respectively. tained from clinical records including registration forms filled out by the pediatric Only 21 cases of RF were observed among 5358 patients with unilateral oncologist at the treating institution, operative notes and pathology narratives after disease who did not have characteristic congenital genitourinary anomalies, nephrectomy, and flow sheets reporting the initial treatment. A few anomalies and their risk was <1%. Although other explanations cannot be completely were ascertained from questionnaires completed by the family. Mention of aniridia excluded, the high rate of RF in patients with the aniridia syndrome chal- on any of these records was taken to be evidence of the WAGR syndrome. Most lenges the view that nephropathy is associated uniquely with missense muta- such patients also had reports of GU anomalies or mental retardation. Whereas it tions in the WT1 gene. It suggests the possibility of a further gradation in the should be emphasized that identification of the 46 patients with WAGR syndrome spectrum of phenotypes associated with different WT1 mutations. Patients was based on clinical criteria only, all 18 of these for whom cytogenetic reports with Wilms’ tumor and aniridia or genitourinary abnormalities should be were available had a deletion or partial deletion at chromosome 11p13. A similar followed closely throughout life for signs of nephropathy or RF. procedure was used for DDS, with the criterion being explicit mention of Drash syndrome or the combination of male pseudohermaphroditism or ambiguous Introduction genitalia with glomerulosclerosis or nephrotic syndrome. This syndrome was not well known during the early years of the study, and it is likely that there was some Wilms’ tumor, an embryonal tumor of the kidney that affects about underascertainment due to failure of the institution to recognize the associated 1 child in every 10,000 in the United States, occasionally occurs as renal pathology (11). A third category of patients consisted of males with hypo- part of the rare WAGR3 (1) or DDS (2, 3) congenital malformation spadias or cryptorchidism who were not already classified as having the WAGR syndromes. Children with the WAGR syndrome invariably have a con- syndrome or DDS. Continuing this hierarchical scheme, patients who had bilateral stitutional chromosomal deletion at 11p13, the location of the WT1 gene disease including metachronous disease in the contralateral kidney were classified (4, 5). Those with DDS usually have a germ-line point mutation, which in the fourth category if they did not already fall in one of the first three. The fifth is predicted to result in an amino acid substitution, in the eighth or ninth and final category consisted of patients with unilateral disease and none of the exon of WT1 (6, 7). It has been suggested that the severe nephropathy congenital anomalies already mentioned. Identification of patients with congenital associated with DDS, which frequently leads to early RF, may result from anomalies or syndromes was based on information available at the time of the the action of altered WT1 in blocking the normal activity of the wild-type Wilms’ tumor diagnosis or shortly thereafter. One patient with no reported anomalies at diagnosis of Wilms’ tumor developed renal pathology characterized by the protein (8). By contrast, because of the less severe genital anomalies and institution as “Drash, nephrotic syndrome” some 20 years after Wilms’ tumor apparent lack of nephropathy associated with WAGR, a reduced WT1 diagnosis and just 2 years before the onset of RF. This patient was not classified dosage during embryogenesis is thought to have a less pronounced effect in the DDS subgroup, however, because the nephrotic syndrome was not evident on development, especially on that of the renal system (6, 7, 9). However, at the time of the Wilms’ tumor diagnosis. exceptions to this model have been observed. Some patients with DDS RF was ascertained as part of the NWTSG Late Effects Study, which also have germ-line deletions predicted to result in truncated WT1 proteins, targeted second malignant neoplasms and congestive heart failure occurring in and one patient with a germ-line missense mutation in exon 9 was free of patients who were treated successfully for Wilms’ tumor. Patients were gen- renal pathology (7). The present study was motivated by the clinical erally followed by their institutions for the first 5–10 years after the Wilms’ observation made by one of us (L. C. S.) of unexpected, late-occurring tumor diagnosis. Thereafter, some continued to be followed by the institution, RF in some patients with the WAGR syndrome. This presented another whereas others were released for direct follow-up through the family or the potential challenge to the idea that nephropathy is associated particularly adult patient by the NWTSG Data and Statistical Center. Approximately 20% with WT1 missense mutations. The large, relatively unselected, NWTSG of patients were lost to follow-up by 10 years, with even higher losses among patient population offered the opportunity to extend and quantify this certain ethnic minorities (12), but this is accounted for in the statistical observation by comparing rates of RF among subgroups of Wilms’ tumor analysis. The criterion for a classification of RF was explicit mention in clinical records or patient reports of chronic RF or end-stage renal disease, with patients who were followed systematically for reasonably long periods of repeated serum creatinine levels above 2.5 mg/dl in patients for whom this time. result was available. Patients with bilateral disease who had surgical removal of both kidneys because of progressive Wilms’ tumor were not counted as Received 10/11/99; accepted 6/12/00. having had RF for purposes of this study, although they had been so counted The costs of publication of this article were defrayed in part by the payment of page in an earlier NWTSG report (13). charges. This article must therefore be hereby marked advertisement in accordance with Statistical Analysis. The cumulative risk of RF was estimated using actuarial 18 U.S.C. Section 1734 solely to indicate this fact. 1 A report from the NWTSG. This research was supported in part by USPHS Grants methods that account for variable follow-up times and losses to follow-up (14). CA54498 and CA42326. The observation time for each subject was the elapsed time from Wilms’ tumor 2 To whom requests for reprints should be addressed, at Department of Biostatistics, diagnosis until the earliest time of RF, death, or last follow-up report through Mail Stop 357232, University of Washington, Seattle, WA 98195-7232. December, 1999. Median follow-up of surviving patients was 11.0 years for the 3 The abbreviations used are: WAGR, Wilms’ tumor, aniridia, genitourinary malformation, and mental retardation; DDS, Denys-Drash syndrome; GU, genitourinary; ILNR, entire cohort, with a range of 10.9–12.2 years over the five subgroups. Of 5201 intralobar nephrogenic rest; NWTSG, National Wilms’ Tumor Study Group; RF, renal patients known to be alive at last contact, 1583 (30%) were followed for 15 or failure. more years, and 598 (11%) were followed for at least 20 years. 4030

Table 1 Incidence of RF by subgroup creased the estimated cumulative risk at 20 years for the WAGR No. of group to 45%; a later onset date would have resulted in an estimated Percent with renal cumulative risk of 69% at 26 years. Table 2 displays key character- Subgroup Patients RFs failure at 20 years istics of the 14 patients with the WAGR syndrome or associated GU DDS 22 13 62.4 anomalies who developed RF. All but four had already undergone WAGR syndrome 46 10 38.3 Male GU anomalies 153 4 10.9 either renal dialysis or transplant by the time the data were compiled. Bilateral disease 397 7 5.5 One WAGR patient had focal glomerulosclerosis present at the time Unilateral disease 5358 21 1.0 of the Wilms’ tumor diagnosis at age 6.9 years but did not develop RF until age 13.6 years. All but one of the patients with DDS who developed RF did so before 9 years of age (Table 3). In contrast, the affected patients with the WAGR syndrome ranged in age from 11.6–28.2 years (median, 14.6 years) at the time of RF. This suggests that the onset of puberty may trigger the events leading to RF in patients with the WAGR syndrome. Eight of the 46 patients with the WAGR syndrome, including 2 of the 10 patients who developed RF, had bilateral disease at onset. The aniridia syndrome is also known to be associated with the precursor lesion known as ILNR (15). A reviewer of an earlier version of this article suggested that treatment of the remaining kidney for bilateral Wilms’ tumor or nephrogenic rests could possibly account for the high rate of RF observed in the WAGR subgroup. Whereas some such explanation cannot be entirely ruled out, it does not seem likely. Review of the available clinical records indicated that none of the patients with the aniridia syndrome were treated surgically for the presence of nephrogenic rests. All 10 patients who developed RF had disease of favorable histology. Five had stage I disease and were treated with dactinomycin and vincristine only. Three had stage II disease (one received dactinomycin and vincristine only, one also received abdominal radiation, and one also received doxorubicin and abdominal radiation). The two patients with bilateral Fig. 1. Cumulative risk of RF by subgroup. (stage V) disease received dactinomycin and vincristine, and one patient also received abdominal radiation. These treatments are no different from those of the vast majority of patients who did not have characteristic Results and Discussion anomalies or syndromes and for whom the rates of RF were substantially The cumulative risk of RF at 20 years ranged from 1–62% over the lower. None of the 10 patients had a relapse of their Wilms’ tumor. five subgroups (Table 1). Fig. 1 graphs the cumulative risk by time The presence of ILNR in patients with unilateral disease but with no since diagnosis. Patients with DDS had the highest risk of early RF, anomalies or syndromes did increase the risk of RF. Restricting attention but rates for those with the WAGR syndrome were higher in later to those diagnosed since 1980 for whom the presence of nephrogenic years; hence, the two curves approached the same level after 25 years. rests was evaluated, 5 cases of RF were observed among 593 patients The jump in the graph at time 0 for patients with DDS reflects the fact who had ILNR, whereas only 3 cases of RF were observed among 2788 that five (23%) of these patients already had evidence of RF at the patients who did not have ILNR (P Ͻ 0.01). The cumulative risk of RF time of diagnosis of their Wilms’ tumor. One patient with the WAGR at 20 years from Wilms’ tumor diagnosis was 3.3% for those with ILNR syndrome died of end-stage renal disease at the age of 27.0 years, and 0.7% for those without ILNR. Because loss or mutation of WT1 is having been diagnosed with Wilms’ tumor at age 1.2 years. However, associated with ILNR, this provides indirect evidence that WT1 mutations because clinical records for the 10 years preceding death could not be may possibly play a role in some cases of RF observed in patients who accessed, the exact time of onset of RF was unknown. For purposes of lack the associated malformation syndromes. the graph, onset was assumed to have occurred 20.8 years after One of 16 female and 9 of 30 male patients with the WAGR syndrome diagnosis of Wilms’ tumor. An earlier onset date would have in- developed RF. The difference is not statistically significant with these

Table 2 Characteristics of patients with the WAGR syndrome or associated GU anomalies who developed RF Sex Age at Wilms’ tumor (yr) Congenital anomaliesa Age at renal failure (yr) Dialysis Transplant Lateralityb Age at last contact (yr) Survival M 1.2 AN/CR/RD 17–27c ? ? U 27.0 Dead/ESRDd M 2.2 AN/CR/RD 28.2 No No U 29.1 Alive M 2.0 HS/RD 21.8 No No LB 23.9 Alive M 0.9 AN/HS 14.9 Yes Yes B 24.9 Alive M 0.6 CR 16.6 Yes No B 21.6 Alive M 1.4 CR 20.1 No Yes U 23.6 Alive M 1.0 AN/CR/HS/RD 17.5 Yes Yes U 22.9 Alive F 2.1 AN/RD 19.0 Yes Yes U 21.2 Alive M 0.8 AN/CR/RD 12.0 Yes No U 12.4 Dead/ESRD M 1.9 AN/CR/RD 14.2 Yes Yes U 18.4 Alive M 5.7 AN 11.8 Yes Yes U 20.4 Alive M 6.9 AN/CR/HS/GS 13.6 No Yes U 17.9 Alive M 1.4 CR 13.1 No No U 15.0 Alive M 1.0 AN/CR/HS 11.6 No No B 11.9 Dead/ESRD a AN, aniridia; CR, cryptorchidism; HS, hypospadias; RD, retardation; GS, glumeruloscerosis. b B, bilateral at onset; LB, late bilateral; U, unilateral. c Exact age unknown; see text. d ESRD, end-stage renal disease. 4031

Table 3 Characteristics of DDS patients who developed RF after the diagnosis of Wilms’ tumor Age at Wilms’ tumor (yr) Age at renal failure (yr) Dialysis Transplant Lateralitya Age at last contact (yr) Survival 1.2 1.2 No No B 1.6 Dead/ESRDb 1.9 2.1 Yes Yes U 14.9 Dead/ESRD 2.9 13.7 Yes Yes U 13.7 Alive 1.6 4.1 Yes Yes U 12.9 Alive 2.1 3.8 Yes Yes U 11.7 Alive 3.8 8.1 Yes Yes U 10.1 Alive 5.0 8.7 Yes Yes U 15.0 Alive 0.8 4.7 No No LB 9.1 Alive a B, bilateral; LB, late bilateral; U, unilateral. b ESRD, end-stage renal disease.

small numbers (P ϭ 0.13). Nonetheless, it is interesting in light of the (Table 1). Thus, the risk of RF in patients whose unilateral Wilms’ finding that germ-line WT1 mutations in Wilms’ tumor patients occur tumor does not occur as part of a known congenital syndrome or in almost exclusively among those who have either DDS or male GU conjunction with one of the characteristic congenital anomalies is anomalies suggestive of an incomplete form of DDS (16). WT1 mutations projected to be exceptionally low, provided that they are successfully are a plausible explanation for the elevated rates of RF seen here (Tables treated with modern front-line chemotherapeutic regimens. 1 and 2) for male patients with GU anomalies who did not have the aniridia or nephropathy associated with the WAGR and DDS, respec- Acknowledgments tively. The four RFs that occurred among male patients with GU anomalies exceed the 1.25 RFs that would have been expected based on rates We thank the many health professionals of the Children’s Cancer Group and among patients without congenital anomalies or syndromes, even after the Pediatric Oncology Group who managed the treatment of these children adjustment for bilaterality (P ϭ 0.01). These patients also received only and without whom the study would have been impossible. standard treatment. Extrapolating from the results of Diller et al. (16), as many as one of four male patients with GU anomalies may carry germ- References line WT1 mutations, most of which would be predicted to result in 1. Miller, R. W., Fraumeni, J. F., Jr., and Manning, M. D. Association of Wilms’s tumor truncated protein. The fact that all four RFs observed among male with aniridia, hemihypertophy and other congenital malformations. N. Engl. J. Med., patients with GU anomalies also took place after the onset of puberty, at 270: 922–927, 1964. ages ranging from 13.1–28.1 years, is consistent with the idea that a 2. Denys, P., Malvaux, P., van den Berghe, H., Tanghe, W., and Proesmans, W. Association d’un syndrome anatomo-pathologique de pseudohermaphrodisme mas- reduced WT1 dosage could be responsible for both their nephropathy and culin, d’une tumeur de Wilms, d’une nephropathie parenchymateuse et d’un mosa- that seen in the patients with the WAGR syndrome. The high rate of RF icisme XX/XY. Arch. Fr. Pediatr., 24: 729–739, 1967. eventually observed among patients with the WAGR syndrome and the 3. Drash, A., Sherman, F., Hartmann, W. H., and Blizzard, R. M. A syndrome of presence of glomerulosclerosis at the diagnosis of Wilms’ tumor in one pseudohermaphroditism, Wilms’ tumor, hypertension and degenerative renal disease. of them suggest that the DDS and WAGR phenotypes are perhaps not so J. Pediatr., 76: 585–593, 1970. 4. Riccardi, V. M., Sunansky, E., Smith, A. C., and Francke, U. Chromosomal imbal- distinct as has been commonly presumed. ance in the aniridia-Wilms’ tumor association: 11p interstitial deletion. Pediatrics, 61: These results suggest the possibility of a gradation in phenotypes 604–610, 1978. associated with WT1 mutations. It starts with the group of patients 5. Gessler, M., Poustka, A., Cavenee, W., Neve, R. L., Orkin, S. H., and Bruns, G. A. having GU anomalies and a moderate long-term risk of RF, progresses Homozygous deletion in Wilms tumors of a zinc-finger gene identified by chromosome jumping. Nature (Lond.), 90: 774–778, 1990. to the group of patients with the WAGR syndrome who have more 6. Pelletier, J., Bruening, W., Kashtan, C. E., Mauer, S. M., Manivel, J. C., Stiegel, J. E., severe GU anomalies and a high long-term risk of RF, and finishes Houghton, D. C., Junen, C., Habib, R., Fouser, L., Fine, R. N., Silverman, B. L., with the group of patients with DDS who have markedly distorted GU Haber, D. A., and Housman, D. Germline mutations in the Wilms’ tumor suppressor development and a high risk of early RF. Sequencing of WT1 for gene are associated with abnormal urogenital development in Denys-Drash syndrome. patients with the male GU anomalies or ILNR who develop late RF Cell, 67: 437–447, 1991. 7. Huff, V. Genotype/phenotype correlations in Wilms’ tumor. Med. Pediatr. Oncol., 27: could help to sort out the correlation between genotype and pheno- 408–414, 1996. type. A weakness of the present report is the fact that little or no 8. Little, M. H., Williamson, K. A., Mannens, M., Kelsey, A., Gosden, C., Hastie, N. D., and information was available regarding the renal pathology that led to van Heyningen, V. Evidence that WT1 mutations in Denys-Drash syndrome patients may end-stage renal disease. Biopsy of the renal lesions in patients in the act in a dominant-negative fashion. Hum. Mol. Genet., 93: 259–264, 1993. WAGR and male GU subgroups is needed to determine whether they 9. Little, M. H., and Wells, C. A. A clinical overview of WT1 gene mutations. Hum. Mutat., 9: 209–225, 1997. involve the mesangial sclerosis typical of that found in the glomeruli 10. Beckwith, J. B., and Palmer, N. F. Histopathology and prognosis of Wilms tumor. of patients with DDS (11). The DDS, WAGR, and GU anomaly Cancer (Phila.), 41: 1937–1948, 1978. subgroups should be followed closely, with regular monitoring of 11. Habib, R., Loirat, C., Gubler, M. C., Niaudet, P., Bensman, A., Levy, M., and Broyer, kidney function and a search for kidney donors in case of signs of RF. M. The nephropathy associated with male pseudohermaphroditism and Wilms’ tumor (Drash syndrome): a distinctive glomerular lesion—report of 10 cases. Clin. Nephrol., The low rate of RF observed in patients who did not have WAGR, 24: 269–278, 1985. DDS, or a characteristic male GU anomaly should be reassuring to the 12. Breslow, N., Olshan, A., Beckwith, J. B., Moksness, J., Feigl, P., and Green, D. vast majority of former Wilms’ tumor patients. Five of the 28 RFs in Ethnic variation in the incidence, diagnosis, prognosis, and follow-up of children with this group were attributed specifically to radiation nephritis and oc- Wilms’ tumor. J. Natl. Cancer Inst., 86: 49–51, 1994. curred among patients enrolled in the first study (NWTS-1) who 13. Ritchey, M. L., Green, D. M., Thomas, P. R. M., Smith, G. R., Haase, G., Shochat, S., Moksness, J., and Breslow, N. E. Renal failure in Wilms’ tumor patients: a report from the received 20 or 30 Gy of abdominal radiation. Five others occurred National Wilms’ Tumor Study Group. Med. Pediatr. Oncol., 26: 75–80, 1996. after the administration of ifosfamide, a known renal toxin predispos- 14. Kaplan, E. L., and Meier, P. Nonparametric estimation from incomplete observations. ing to Fanconi’s syndrome, for treatment of relapse among patients J. Am. Statist. Assoc., 53: 456–481, 1958. enrolled in the third or fourth studies (NWTS-3 and -4). Elimination 15. Beckwith, J. B., Kiviat, N. B., and Bonadio, J. F. Nephrogenic rests, nephroblasto- matosis, and the pathogenesis of Wilms tumor. Pediatr. Pathol., 10: 1–36, 1990. of these 10 events reduced the estimated rates of RF at 20 years from 16. Diller, L., Gharemani, M., Morgan, J., Grundy, P., Reeves, C., Breslow, N., Green, Wilms’ tumor diagnosis from 5.5% to 4.5% among those with bilat- Neuberg, D., Pelletier, J., and Li, F. P. Constitutional WT1 mutations in Wilms’ tumor eral disease and from 1.0% to 0.6% for those with unilateral disease patients. J. Clin. Oncol., 16: 3634–3640, 1998.

4032

Norman E. Breslow, Janice R. Takashima, Michael L. Ritchey, et al.

Cancer Res 2000;60:4030-4032.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/60/15/4030

Cited articles This article cites 14 articles, 1 of which you can access for free at: http://cancerres.aacrjournals.org/content/60/15/4030.full#ref-list-1

Citing articles This article has been cited by 6 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/60/15/4030.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/60/15/4030. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.