| |||||||||||||||||||||||||||||||
Cytologic examination of urine was introduced by Papanicolaou and Marshall in 1945, and it has been successfully used to screen for carcinoma of the bladder3. Although it can detect some preclinical cancers, urinary cytologic examination has a number of limitations. False negative results can be obtained when the specimens are scanty, when the tumors are low grade, and when diagnostic cells are masked by larger numbers of non-neoplastic cells3,4,5,6,7. Furthermore, the success of the technique depends heavily on the expertise of the cytopathologist.
Many of the shortcomings of urinary cytologic examination can be overcome with molecular biologic techniques, particularly the polymerase chain reaction (PCR). Because of their remarkable sensitivity, PCR-based techniques have been used to detect exfoliated neoplastic cells in the stool of patients with colorectal neoplasms8; in the sputum of patients with lung cancer9; in the blood, bile, and stool of patients with pancreatic cancer10,11 (and Caldas C: unpublished data); and in the urine of patients with bladder cancer12. We applied one of these PCR-based techniques to archival tissue from the late Hubert H. Humphrey12.
Case Report
To understand why we chose Humphrey as an example, one must be familiar with his medical history13,14,15,16,17. In May 1967, while he was vice president of the United States, he was admitted to Bethesda Naval Hospital for hematuria. Voided urine specimens were obtained for cytologic examination, and cystoscopy was performed. The condition of his bladder was found to be grossly consistent with chronic proliferative cystitis, and a biopsy revealed only a minute focus of dysplastic change in the transitional epithelium. No cancer was identified grossly by inspection or in the biopsy specimens that were taken. A number of pathologists reviewed the urine-cytology slides (Figure 1), among them Dr. John K. Frost, director of cytopathology at Johns Hopkins Hospital. Frost believed that the cytology slides were, in fact, diagnostic of carcinoma. Other experts apparently did not agree, and a definitive diagnosis was not established. Humphrey was therefore followed with cystoscopy every six months13. A biopsy revealed in situ carcinoma in 1969, but he was asymptomatic until four years later. In 1973 a biopsy of his prostatic urethra revealed a "borderline malignancy" (in situ transitional-cell carcinoma with a small focus of probable microinvasive cancer), for which he received both radiation therapy and intravesicular thiotepa13. In August 1976 recurrent hematuria developed, and a biopsy led to the diagnosis of infiltrating carcinoma of the bladder. A radical cystectomy was performed, revealing widely infiltrative transitional-cell carcinoma of the bladder with lymph-node metastases16,18,19. Humphrey died of the cancer on January 13, 1978.
|
With the permission of Humphrey's widow, we obtained the formalin-fixed, paraffin-embedded tissue blocks of the infiltrating bladder cancer resected from Humphrey in 1976 and screened the cancer for mutations in the p53 tumor-suppressor gene20. Five-microm sections of this primary carcinoma were microdissected to enrich for neoplastic cells, and exons 5 to 9 of the p53 gene were amplified by PCR12. The PCR products were then isolated and subcloned into bacteriophages, and more than 500 pooled clones were sequenced,12 revealing a transversion from adenine to thymine in codon 227 of the p53 gene in the primary carcinoma.
Next, we obtained two of the Papanicolaou-stained filters (Millipore, Bedford, Mass.) that had been prepared with Humphrey's urine when he presented in May 1967. We isolated DNA from these filters and amplified by PCR a region of 104 base pairs encompassing codon 227 of the p53 gene. The amplification primers for this portion of exon 7 of the p53 gene were 5'GTAGGAATTCCAAGGCGCACTGGCCTC3' and 5'AGGAATTCTTACACATGTAGTTGTAGTGG3'. The PCR conditions and the methods used for cloning into lambda Zap (Strategene, La Jolla, Calif.) were performed as previously described12. We then cloned the PCR products into the bacteriophage vector, transferred them to nylon membranes, and hybridized them with a 32P-labeled oligonucleotide probe (5'GTTGGCTCAGACTGTACC3') specific for the codon 227 mutation found in the resected primary carcinoma12.
Results
Both the invasive bladder carcinoma resected in 1976 and the filters prepared from urine in 1967 were analyzed for p53 mutations. Sequencing of pooled clones of DNA from the cancer resected in 1976 revealed a transversion from adenine to thymine in codon 227 of the p53 gene (Figure 2A). This mutation creates a cryptic splice site in exon 7 of the p53 gene21. Normal splicing of introns permits the correct assembly of exons and the production of a full-sized, functional protein. The hidden splice site created by the mutation in codon 227 results in the loss of several amino acids and in the production of a shortened, mutant p53 protein. This mutation was not present in non-neoplastic tissue obtained from the muscularis propria of Humphrey's resected bladder.
|
Discussion
The detection of cells harboring a p53 mutation in Humphrey's urine-cytology specimen from May 1967 has a number of implications. First, although clonal p53 mutations have not been identified in non-neoplastic urothelium, they have been found in approximately 60 percent of bladder carcinomas12,22,23,24. In addition, the mutation in the DNA obtained from Humphrey's urine-cytology specimen was the same as the one identified in his primary carcinoma. These observations suggest that the neoplasm was already present in May 1967. Second, the presence of cells harboring p53 mutations in the urine suggests that, if the importance of the mutation had been known in 1967 and the technology described here had been available, Humphrey's neoplasm could have been detected then. Third, p53 mutations and the associated overexpression of the p53 gene product are generally, but not universally, associated with in situ lesions that tend to progress and with high-grade or advanced-stage carcinomas12,25,26,27,28,29,30. This raises the possibility that Humphrey's neoplasm was already in a phase of aggressive growth in 1967. Had Humphrey known that he had aggressive bladder cancer in 1967, he might have withdrawn from the presidential race14. More important, he and his physicians might have opted for more aggressive, and potentially lifesaving, surgery years earlier.
Although the case presented here is quite dramatic, the techniques we used may not be ideal for screening the general population for bladder cancer. First, p53 mutations may occur late in the course of some bladder cancers12,25,26,27,28,29,30. Second, mutations in p53 are not limited to a single codon, but can involve a number of different sites in the gene, making screening tests based on the detection of p53 mutations technically challenging. Third, approximately 40 percent of bladder cancers do not harbor p53 mutations, and such cancers would not be detected with these techniques. Despite these limitations, the success of PCR-based techniques in this case reinforces the recognition of their powerful potential for the early detection of cancer. For this potential to be fully realized, further studies must define the molecular events responsible for the development of carcinoma of the bladder. The identification of a putative tumor-suppressor gene on chromosome 9 may provide a better molecular test for the early detection of this neoplasm31,32,33,34,35.
Supported in part by a collaborative research arrangement with Oncor, Inc., Gaithersburg, Md.
We are indebted to Shirley Myers, Norman Barker, Claire E. Hruban, Dr. Victor Reuter, Dr. Ted Ganster, Charles Desaulniers of the Millipore Corporation, and Muriel Humphrey-Brown for their assistance.
Source Information
From the Departments of Pathology (R.H.H., Y.S.E.) and Otolaryngology-Head and Neck Surgery (R.H.H., P.R., D.S.), Johns Hopkins Medical Institutions, Baltimore. Presented in part at the annual meeting of the U.S. and Canadian Academy of Pathology, San Francisco, March 16, 1994.
Address reprint requests to Dr. Sidransky at the Department of Otolaryngology-Head and Neck Surgery, 818 Ross Research Bldg., 720 Rutland Ave., Baltimore, MD 21205-2195.
References
| |||||||||||||||||||||||||||||||
Related Letters:
Hubert Humphrey's Bladder Cancer
Homer R. J., Braund W.J., Hruban R. H., van der Riet P., Sidransky D.
Extract |
Full Text
N Engl J Med 1994;
331:880-881, Sep 29, 1994.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
