Donor-derived bladder cancer following kidney transplantation

Abstract

Kidney transplantation is associated with elevated cancer risk. We report a 75- year-old woman diagnosed with bladder cancer five years after receiving a kidney transplant from a male donor. Whole genome sequencing of this tumor revealed many DNA reads that mapped to the Y chromosome, and fluorescent in situ hybridization identified that 32% of tumor cells had a Y chromosome. These results support a donor origin of this cancer, with the most plausible source being the urothelium of the donor allograft. Sequencing also identified BK polyomavirus, which likely contributed to the development of this cancer.

1. Introduction

Kidney transplantation is the treatment of choice for end-stage kidney disease but requires the life-long administration of immunosuppressive therapy to prevent rejection. Recipients of transplanted kidneys or other organs experience an elevated risk of cancer,^1,2 in part related to loss of immune surveillance against precancerous lesions and loss of immune control of oncogenic viruses.

In particular, solid organ transplant recipients have an elevated risk of bladder cancer 1, 2, 3. Recent research provides evidence for a role for BK polyomavirus (BKPyV) in the etiology of a subset of posttransplant bladder cancer 4, 5, 6. BKPyV infection is almost universal in adults and usually asymptomatic, but viral reactivation among immunosuppressed kidney recipients can lead to BKPyV nephropathy, a major cause of graft loss.⁷

Since kidney recipients retain their native urinary bladder after transplantation, it is expected that the vast majority of bladder cancers in this setting are derived from the recipient's cells. Of interest in this regard, two case reports have described bladder cancers in kidney recipients that apparently arose from donor cells.^8,9 These reports prompted us to reevaluate a series of posttransplant bladder cancers that we previously studied⁵ for evidence of donor origin.

2. Case presentation

Because this case report relies on information from transplant and cancer registries, clinical details are very limited. We have also rounded the ages of the kidney recipient and donor to preserve confidentiality. This study was approved by the institutional review board at Cedars Sinai Medical Center.

In our prior study,⁵ we used linked data from the United States solid organ transplant registry and multiple state cancer registries to identify bladder cancers arising in transplant recipients. We obtained formalin fixed paraffin-embedded (FFPE) tumor tissues from 43 of these bladder cancers and a smaller number of normal tissue samples adjacent to the tumors. DNA and RNA were successfully extracted from 38 tumors, and library preparation and whole genome and RNA sequencing were performed as described previously.⁵ For this study, DNA reads mapping to the X and Y chromosomes were counted and normalized to the total reads and length of the respective chromosome to get reads per kilobase per million (RPKM). Median values of normal samples were used to calculate the relative copy numbers of X and Y chromosomes in each tumor.

As shown in Fig. 1, most bladder cancers and normal tissues from males and females showed the expected number of copies of the X and Y chromosomes. Five tumors were estimated to contain only a single X chromosome and no Y chromosome (i.e., loss of X or Y, corresponding to an XO karyotype), while seven tumors from males showed extra copies of X and/or Y, suggesting chromosome duplication.

Fig. 1.

Panel A. Scatterplot of chromosome Y versus chromosome X copy numbers in bladder tumors and available normal tissues from solid organ transplant recipients, estimated from DNA whole genome sequencing and annotated by recipient sex and inferred ploidy.

Panel B. Scatterplot of chromosome Y versus chromosome X DNA reads per kilobase per million (RPKM) in bladder tumors and available normal tissues from female solid organ transplant recipients.

Panel C. DNA coverage for chromosome Y in case TBC07. The pseudoautosomal region is highlighted in green, centromere in red, and G banding in gray. In the diagram, the right side of the Y chromosome corresponds to a region with highly repetitive sequences, and DNA reads could not be reliably matched to this region.

Notably, one bladder cancer (TBC07) from a female kidney recipient demonstrated a noticeably higher number of reads mapping to the Y chromosome than observed for the other females, although the estimated the number of copies of the Y chromosome in her tumor was lower than for the XY tumors among males (Fig. 1A and B). A large number of reads (n = 858,986) for this tumor mapped to the pseudo-autosomal region of Y, which has homology to the X chromosome and which causes a slight positive trend towards more apparent Y copies with increasing X copies in other female patients (Fig. 1B). Nonetheless, many reads (n = 1,596,317) mapped to the unique mappable regions of the Y chromosome (Fig. 1C).

This patient was a 75-year-old woman diagnosed in 2010 with an undifferentiated papillary transitional cell carcinoma of the bladder that presented at a localized stage (Fig. 2A and B). Five years earlier, she had received a deceased-donor kidney transplant for renal disease of unknown cause. The donor was a male who was <20 years old at the time of his death from head trauma and had no history of cancer. The patient received surgical treatment of an unspecified nature for her bladder cancer. The subsequent status of her cancer is unknown, and she died of pneumonia approximately two years after cancer diagnosis. The recipient and donor were matched at the human leukocyte antigen A, B, and DR loci, precluding us from using targeted sequencing of this region of the tumor genome to distinguish between recipient and donor origin.

Fig. 2.

Panel A. On the digitally scanned H&E image, the tumor areas of the bladder specimen are annotated in red and green.

Panel B. At high power, the invasive high grade urothelial carcinoma is composed of irregular nests of cells with marked nuclear atypia, prominent nuclei, and mitotic figures.

Panel C. Scanned in situ hybridized image showing DAPI stained tumor area was matched to H&E image and annotated green.

Panel D. FISH results using an XY combination probe (MetaSystems, Medford, Massachusetts) where the X chromosome is green and the Y chromosome is orange. The GATA3 probe (Empire Genomics, Depew, New York) is shown in red (CY5). Far left cell: normal female cell (XX). Center cell: male tumor cell (XY GATA3+). Right cell: female tumor cell (XX GATA3+).

We performed fluorescence in situ hybridization (FISH) on an FFPE section of the tumor from this patient, using probes for the X and Y chromosomes (Metasystems, Medford, Massachusetts) and the tumor cell marker GATA3 (Empire Genomics, Depew, New York). One hundred cells were scored across two regions of the tumor, of which 74 were XX consistent with female origin (36 of these were also GATA3+, indicating they were tumor cells, and 38 were GATA-), 16 were XY indicating male origin (all of which were GATA3+), and the remaining 10 cells were GATA3+ but exhibited aneuploidy (6 were X0 and 4 were XXYY). Thus, of the 62 GATA3+ tumor cells, 20 (32%) had at least one Y chromosome. Representative fluorescent images are shown in Fig. 2C and D.

As described previously,⁵ we also performed metagenomic analysis of both whole genome sequencing and RNA sequencing data from these bladder cancers. Our analysis of this patient's tumor identified reads mapping to two viral genomes (Fig. 3A). BKPyV (NC_001538.1) was detected with near-complete genome coverage in the whole genome sequencing data (97.6% of the 5153 bp viral genome), accompanied by an RNA sequence coverage of 0.31 RPKM, indicating some active transcription (Fig. 3B). However, BKPyV large T antigen expression could not be confirmed by immunohistochemistry (pab416 antibody, Sigma-Aldrich, St. Louis, Missouri).

Fig. 3.

Panel A. Breadth and normalized depth in reads per kilobase per million (RPKM) for BK polyomavirus (BKPyV) and Epstein-Barr Virus (EBV) in DNA and RNA sequencing from tumor TBC07.

Panel B. Depth of coverage for RNA (red) and DNA (gray) for BKPyV from tumor TBC07, with viral genes indicated below the x-axis.

Panel C. Depth of coverage for RNA (red) and DNA (gray) for EBV from tumor TBC07, with open reading frames indicated below the x-axis (sense direction, blue; antisense direction, orange).

Epstein-Barr virus (EBV; NC_007605.1) was also detected in this tumor, but at markedly lower levels, with only 11.5% genome coverage in the whole genome sequencing data and negligible RNA expression (RPKM 0.004), suggesting low-level infection without significant transcriptional activity (Fig. 3C).

3. Discussion

Two prior reports of bladder cancer of donor origin arising in a kidney recipient provide details that complement our case findings.^8,9 Michel Ortega et al. described a 69-year-old male who received a kidney from his sister.⁸ Six years later, the patient developed BKPyV nephropathy. Approximately eight years after transplantation, he developed hematuria and was diagnosed with high grade invasive bladder cancer and incidental prostate cancer. Treatment included radical cystectomy and bilateral removal of the native kidneys and ureters with creation of an ileal conduit. Two years later, the patient was diagnosed with a urothelial carcinoma arising in the upper urinary tract of the donor kidney. Testing of DNA from the upper urinary tract cancer using a genotyping microarray revealed that 50% of the cells had an XY genotype and 50% had an XX genotype, and a FISH image of the same tumor demonstrated tumor cells with XX or XXXX karyotype. The bladder cancer showed only cells with an XX karyotype on FISH (Daynna Wolff, personal communication, July 7 2025), matching the sex of the donor.

In the second case report, Qian et al. described a 56-year-old male who received a kidney transplant from a 51-year-old male deceased donor.⁹ His posttransplant course was also complicated by BKPyV nephropathy. Seven years posttransplant, evaluation for decreased renal function revealed multifocal muscle-invasive bladder cancer. Nine months after radical cystoprostatectomy, the patient presented with gross hematuria, leading to diagnosis of a urothelial carcinoma involving the donor kidney and ureter. Short tandem repeat (STR) testing of DNA from the bladder and allograft tumors revealed both to be of donor origin.

Like the case reported by Michel Ortega,⁸ evidence for the donor origin of the bladder cancer in our recipient comes from the sex mismatch between her and her male donor. Specifically, we detected Y chromosome positive tumor cells by whole genome sequencing and FISH. Of interest, only a fraction of the tumor cells had a Y chromosome karyotype—approximately 32% by FISH—while 68% of tumor cells showed an XX or X0 karyotype. Only tumor cells (GATA3+ on FISH) contained a Y chromosome.

The plausible source of donor-origin bladder tumor cells in these three cases is the urothelium of the donor allograft. Both prior patients also had upper urinary tract carcinoma in the donor kidney.^8,9 Although diagnosed after the bladder cancers, these lesions were likely present earlier, with bladder tumors arising via seeding from the upper tract. Tumor spread from the upper urinary tract could occur through urine, urothelial continuity, or hematogenous routes, consistent with mechanisms proposed for some bladder cancers arising in the general population.¹⁰ Our patient lacked a documented upper urinary tract tumor, but an undetected lesion could have been the source. Furthermore, one might expect that, in the setting of kidney transplantation, such tumors would arise at the junction of the donor ureter and bladder. In both prior reports,^8,9 the bladder cancers were multifocal, whereas information on the location within the bladder of our cancer case was unavailable. Verine et al.¹¹ describe a renal papillary carcinoma in a donor kidney that showed recipient origin, suggesting similar cross-seeding of cancer within the urinary tract.

Although FISH supports donor-derived cells in our tumor, these were a minority, with most cells showing an XX or X0 karyotype. One explanation for this mixed cell population is heterogeneous Y chromosome loss, observed in 10–40% of male bladder cancers and linked to aggressive behavior.¹² Alternatively, the cancer could represent a “collision” of two independent tumors—one donor-derived and one recipient-derived—as described in other contexts.¹³ A third possibility is that our case patient's bladder cancer arose primarily from her native bladder cells and that donor stem cells, transiting to the site of the developing tumor, contributed in a secondary capacity, as has been proposed in several reports of transplant-associated skin cancer.^14,15 One of the previously reported cases of donor-derived bladder cancer had only donor-derived cells visible on FISH⁸ (Daynna Wolff, personal communication), while the other report did not describe the fraction of donor vs. recipient cells in the tumor.⁹

Finally, we emphasize the possible etiologic role of BKPyV in our bladder cancer case and the two previous cases.^8,9 A growing body of evidence supports the involvement of BKPyV in posttransplant bladder cancers 4, 5, 6. Both prior patients,^8,9 before their cancer diagnosis, had developed BKPyV nephropathy, which is associated with immunosuppression, high-level BKPyV replication, and urothelial infection. In the setting of immunosuppression, persistent BKPyV infection of bladder cells may lead to malignant transformation.⁶ Our case lacked documented BKPyV nephropathy, but her tumor was BKPyV + based on genomic coverage and viral transcription; immunohistochemical staining for BKPyV large T antigen was negative, but this method may be less sensitive than sequencing (as has been shown for a related virus, Merkel cell polyomavirus).¹⁶ In all three cases, BKPyV may have been important for carcinogenesis, particularly given the presence of an initially healthy donor urothelium. Indeed, the donor in our case was less than 20 years old—an age when urothelial cancer is rare—making it unlikely that he had any meaningful preexisting damage in his upper urinary tract.

In conclusion, our case report and the two published previously^8,9 demonstrate that donor-derived cells (probably from the upper urinary tract) appear to contribute to some bladder cancers arising among immunosuppressed kidney transplant recipients. How frequently this occurs, and whether a similar phenomenon may underlie a subset of bladder cancers in the general population,¹⁰ are open questions. Our findings also add support for a role for BKPyV in posttransplant bladder cancer 4, 5, 6.

CRediT authorship contribution statement

Eric A. Engels: Writing – original draft, Supervision, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Sandra Burkett: Writing – review & editing, Investigation, Data curation. V Krishnan Ramanujan: Writing – review & editing, Investigation, Data curation. Kelly J. Yu: Writing – review & editing, Investigation, Data curation. Petra Lenz: Writing – review & editing, Investigation, Data curation. Aaron M. Rozeboom: Writing – review & editing, Investigation, Data curation. Gabriel J. Starrett: Writing – original draft, Resources, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Ethics declaration

The authors have NOT obtained informed consent from participants or their legal representatives. The study used only archived tumor tissue and information collected in transplant and cancer registries. Consent from the patient and family was not required.

This study included organ or tissue donors. This study includes human biological material that was left over from clinical testing. The IRB determined that consent was not required. The samples used in this research were not sourced from executed prisoners or prisoners of conscience.

This study was performed in compliance with relevant laws, regulatory frameworks and guidelines where the research took place. This study was approved by the Cedars Sinai Health Sciences University. (Approval No. n/a)

This research follows the CARE guidelines and the CARE checklist.

Data sharing

All sequencing data used in this study is available from dbGaP under accession phs003012.v1.p1 or from supplemental data in Starrett et al..⁵

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.