ALLELIC LOSS OF 8p SEQUENCES IN PROSTATIC lNTRAEPlTHELlAL NEOPLASIA AND CARCINOMA MICHAEL
KIRK J. WOJNO,
JILL A. MACOSKA
ABSTRACT Objectives. Previous work has suggested that prostatic intraepithelial neoplasia (PIN) may be a premalignant lesion important in tumorigenesis of the prostate. However, to adequately test this hypothesis at the genetic level, it is necessary to determine whether lesions in close proximity demonstrate similar genetic alterations and, hence, whether an “evolutionary” relationship might exist between PIN and tumor in the same prostate. Therefore, the purpose of this study was to examine at least two PIN lesions per prostate (one adjacent to and another distant from malignant lesions in the same prostate) for similarities or differences in the types and frequencies of genetic alterations. Methods. To accomplish this goal, DNA was extracted from microdissected PIN, tumor, and normal epithelial tissue samples from 48 radical prostatectomies and amplified using polymerase chain reaction techniques at highly polymorphic microsatellite repeat sequences at proximal (D8S87, 8~12) and distal (NEFL, 8~21) loci on the short arm of chromosome 8. PIN specimens were either adjacent to (within one high-power microscopic field [HPF]) or distant from (separated by two or more HPFs) tumor specimens from the same patients. Results. Similar fractional allelic loss frequencies were observed for informative tumor (10 [35%] of 29) and PIN (6 [2 1%] of 29) samples at the NEFL locus, but allelic loss at the D8S87 locus was observed only in tumors (8 [22%] of 36 informative samples). Moreover, allelic loss at the NEFL locus involved the same allele in 4 cases and different alleles in 3 cases. Interestingly, all 4 cases with the same allele losswere from adjacent PIN and tumor tissues, and all 3 with different allele loss were from distant PIN and tumor. Conclusions. These results suggest that PIN and invasive cancer share common genetic events (eg, deletion at the NEFL locus) along the same pathway of development in the prostrate. UROLOGY 50: 643-647, 1997. 0 1997, Elsevier Science Inc. All rights reserved.
rostatic intraepithelial neoplasia (PIN) is considered the most likely precursor of prostate cancer (PCs).’ High-grade PIN is cytologically similar to cancer and can be diagnosed by features such as the size and architectural configuration of the glands, enlarged and hyperchroThis study was supported by NIHINCI Grant CA60948 U.A.M.1, funds from the Michigan Prustute Institute (M.J.H. and J.A.M.), grants from the Swedish Medical Research Council (MJ.H.1, the Swedish Medical Association (M.J.H.), Lion’s Foundation (M.J.H.), thejohanna Hagstrand and Sigfrid Linnkr Foundation (M.J.H.), and the K.G. Lennander Foundation (M.J.H.1 From the Department of Surgery, Section of Urology, the Department ofpathology, and The Michigan Prostate Institute, The University of Michigan Medical Center, Ann Arbor, Michigan Reprint requests: Jill A. Macoska, Ph.D., Department of Surgery, Section of Urology, The Michigan Prostate Institute, The University of Michigan, 5510 MSRB I, 1150 West Medical Center Drive, Ann Arbor, Ml 48109-0680 Submitted:]anuary 23, 1997, accepted (with revisions): April 23, 1997 0 1997, ELSEVIER SCIENCE ALL RIGHTS RESERVED
matic nuclei, macronucleoli, and disruption of the basal cell layer.lx2 Immunohistochemical analyses of biochemical markers, such as HERZ/neu, epithelial growth factor-receptor, type IV collagenase, and transforming growth factor-alpha, have consistently placed high-grade PIN in the continuum between normal prostate and malignant epithelium (see Bostwick3 for review). Like carcinoma, PIN is responsive to androgens, and androgen-deprivation therapy decreases the prevalence and extent of high-grade PIN.4 Initial findings of high-grade PIN on needle biopsy have proved predictive of carcinoma in one-third to one-half of subsequent biopsies.5z6 Moreover, these carcinomas are adjacent to the original PIN site in 65% of cases.’ In some cases, PIN foci have been observed budding from microcarcinomas via morphologically intermediate transitive glands.’ PIN and carcinoma volumes are inversely correlated,’ and autopsy studies have 0090.4295/97/$17.00
PII soogo-4295(97)oo3o4-x 643
shown that high-grade PIN chronologically precedes PCs by at least 10 years.“,” Taken together, these observations suggest that prostate tumor progression may involve conversion of PIN lesions into invasive carcinomas. Molecular genetic studies have shown that deletion of sequences mapping to the short arm of chromosome 8 (8~) and gain of the long arm of chromosome 8 (Bq) occur with equivalent frequencies in high-grade PINS and carcinomas.‘2-‘4 This finding suggests that PIN and PCs are linked at the genetic level and that a subset of PIN lesions may progress to PCs. However, these studies did not attempt to correlate the spatial proximity of PIN and PCs lesions with their respective genotypes. Examination of the spatial relationship between genetically altered malignant and PIN lesions is necessary to determine whether lesions in close proximity demonstrate similar genetic alterations and, hence, whether an “evolutionary” relationship might exist between PIN and tumor in the same prostate. Therefore, the purpose of the present study was to examine at least two PIN lesions per prostate (one adjacent to and another distant from malignant lesions in the same prostate) for similarities or differences in the types and frequencies of genetic alterations. To accomplish this aim, retention or loss of specific alleles of highly polymorphic microsatellite sequences at 8p loci were examined in normal, PIN, and malignant tissues from 48 cancerous prostates. The results of this study show that adjacent PIN and tumor lesions tend to share identical genetic alterations of Bp, consistent with the identification of PIN as a premalignant lesion for invasive carcinomas of the prostate. MATERIAL
TISSUE ACQUISITION AND CHARACTERIZATION Prostate tissue was obtained after radical prostatectomy from 53 patients diagnosed with PCs. Tumor pathologic stage was assigned using a modification of the TNM system, as described by a consensus paneli Degree of tumor differentiation (grade) was assigned according to the combined Gleason score.i6 This information is summarized in Table I. Preoperative prostate-specific antigen (PSA) values were available for 46 patients.
TISSUE MICRODISSECTION After pathologic evaluation of radical prostatectomy tissue, paraffin-embedded tumor specimens were serially sectioned at 4 pm. These specimens comprised areas of at least 70% malignant cells as well as adjacent (within one high power field [HPF], equivalent to 400X magnification) or distant (at or farther than two HPFs) high-grade PIN lesions and normal or hyperplastic epithelium. Archival tissues were used rather than frozen specimens to ensure accurate histologic identification of PIN lesions. One section of each series was stained with hematoxylin/eosin to define areas of discrete histology, which were circled on the slide (K.J.W., C.P.P.). The remain-
Pathologic and grade
Grade 5, 6 7 8, 9
15 16 0
3 9 3
1 1 0
19 26 3
ing adjacent nonstained step sections were deparaffinized with xylenes, the circled areas were separated from surrounding tissues by needle dissection, and the DNA was extracted as previously described.i7,1R
PURIFICATION To prepare DNA for amplification, circumscribed tissue areas were scraped into 0.5-mL tubes and digested in 1~ polymerase chain reaction (PCR) buffer (50 mM KCl, 10 mM TrisHCl, pH 8.3, 2.5 mM MgClz) supplemented with 0.1 mg/mL of proteinase K. A 2-mm X 2-mm tissue area was typically incubated in 10 PL of digestion buffer; larger tissue areas were digested in proportionately larger volumes. The samples were incubated for 12 to 16 hours at 37°C with shaking, then for 20 minutes at 95’C to inactivate the proteinase K. DNA from normal tissue and PIN and tumor lesions was successfully extracted in 48 of the 53 cases chosen for study and was analyzed by PCR (as described below).
GENETIC ANALYSIS PCR amplification assays targeted sequences containing highly polymorphic microsatellite repeat markers NEFL at 8~21 (Rogaev et al.‘“) and D8S87 at 8~12 (Weber et al.“‘). Primer sequences utilized for NEFL were (forward) 5’ GCAGTAGTGCCGCAGTTTCA 3’; (reverse) 5’ TGCAATTCATCTTCCTTTCT 3’; for D8S87, the sequences were (forward) 5’ GGGTTGTTGTAAATTAAAAC 3’; and (reverse) 5’ TGTCAAATACTTAAGCACAG 3’. These loci were chosen because they map to a region (8~12-21) frequently deleted in PINS and carcinomasi For each PCR reaction, 5 PL of DNA was amplified in a 20-PL mixture comprising 200 PM each of deoxyguanosine triphosphate, deoxyadenosine triphosphate, deoxythymine triphosphate, and deoxycytidine triphosphate (dCTP); 1 X PCR buffer (50 mM KCl, 10 mM Tris-HCl, pH 8.3, 2.5 mM MgC12); 1 PM oligonucleotide primers; 0.6 U of Taq Polymerase (Life Technologies), and 100 ng of oligonucleotide primer end-labeled with -y32-PdCTP. The reactions were heated to 95°C for 5 minutes (to completely denature the template) and were then cycled at 95°C for 1 minute, 55°C for 1 minute, and 72°C for 1 minute for 35 cycles; 72°C for 7 minutes; and 4°C to stop the reactions. Aliquots of each reaction were electrophoresed on 6% acrylamide/‘l M ureasequencing gels, and the gels were autoradiographed. Allelic loss was scored when the ratio of allelic signal intensities in tumor tissue was less than or equal to 50% of that for the same alleles in normal tissue from the same heterozygous patient by two independent observers (M.J.H., J.A.M.). Cases of borderline loss of heterozygosity (LOH) were also evaluated by densitometry (IS-1000 Digital Imaging System, Alpha Innotech Corporation). Instances of apparent LOH were repeated for verification. Only those samples that exhibited LOH in subsequent amplification reactions were considered positive for allelic loss. UROLOGY
SO (4), 1997
7 9 12 24 26 27 32 34 37 38 39 48 51
6 7 7 6 7 7 7 9 7 8 7 7 7
T2 T2+ T3a+ T3a T3b T2 T4+ T3c T3b+ T3b+ T3c+
LL LU LU LL R LU LU LU LU R NI
R I LU LL LL LU R LU R R NI R NI
NEFL T2 R
LU LL LU LU
and allelic loss patterns
LU R R LU R R R LU R R NI R NI
R R R R R R R R R R NI R NI
LL LU R R R NI R LU LU LU LU LU LU
R R R R R NI R R R R R R R
D8S87 T2 R
R NI LU LU
R R R R R NI R R R R R R R
R R R R R NI R R R R R R R
KEY: Tl and T2 = distinct tumorfoci greater than two high-powerfields (HPFs) distantfrom each other within the prostate; Pl and P2 = prostatic intraepithelial neoplasia lesions within one HPF of Tl and T2, respectively; N = normal prostatic epithelium; LL = loss of lower allele; LLJ = loss of upper allele; R = heterozygous, both alleles retained; NJ = homozygous, not informative; I = microsatellite Instability, an expansion ofthe number ofmicrosatellite repeats apparentfor the upper allele of this specimen.
STATISTICAL ANALYSIS Statistical analysis was performed (Fisher’s exact) test. A P value <0.05 tically significant.
utilizing a chi-square was considered statis-
RESULTS CORRELATIONOFALLEL~CLoss PATTERNSWITH PATH~L~G~CAND CLrNrcAL PARAMETERS Thirteen of the 48 tumors examined demonstrated allelic loss at NEFL (5 tumors), D8S87 (3 tumors), or both loci (5 tumors), as detailed in Table II. Allelic loss frequencies were 35% (10 of 29 informative tumors) at NEFL and 22% (8 of 36 informative tumors) at D8S87. Eleven of the 13 tumors with allelic loss had a Gleason score of 7, 8, or 9 (11 [38%] of 29 with a total grade 7, 8, or 9) compared with 2 tumors that had a score of 5 or 6 (2 [ll%] of 19, P = 0.46). Ten tumors with loss extended outside the prostate (10 [59%] of 17 Stage T3/T4), whereas 3 were localized to the prostate (3 [lO%] of 31 Stage T2, P = 0.011). There were no significant differences in stage or grade between tumors with allelic losses at NEFL versus D8S87. These data are summarized in Table II. The mean (&SD) preoperative PSA values were 13.82 + 23.65 for patients with and 6.45 t 3.59 for those without tumor allelic loss. This difference approached statistical significance (P = 0.080). ALLELIC Loss PATTERNSAT NEFLLOCUSARE SIMILARINPINAND PCA Six cases,with a total of seven PIN lesions, demonstrated allelic losses at the NEFL locus (6 [21%1 of 29 informative lesions) versus none at the D8S87 locus (0 of 36 informative lesions, P = 0.031). Four cases demonstrating the same allele deleted in PIN and tumor lesions (12, 24, 27, and 34) involved adUROLOGY
50 (41, 1997
TABLE III. Summary of allelic loss patterns in tumor and prostatic intralesional neoplasia NEFL Same allele, adjacent Different allele, adjacent Same allele, distant Different allele, distant Loss in tumor only Loss in PIN only Total KEY: PIN = prostatic
4 0 0 3 4 0 11
D8S87 0 0 0 0 8 0 8
jacent tissues (tumor and PIN within one HPF), whereas 3 cases(7,24, and 26) with different alleles deleted in PIN and tumor involved distant tissues (tumor and PIN separated by at least two HPFs, P = 0.013) (Table III). This correlation was observed even within a single prostate (case 24), where adjacent tumor and PIN lesions shared loss of the same allele at NEFL, whereas the distant PIN lesion demonstrated loss of the other allele at the same locus (Fig. 1). The pathologic stages of tumors associated with genetically normal or abnormal PIN lesions were similar. Examples of allelic loss in adjacent and distant PIN and tumor foci are shown in Figure 1.
COMMENT The present data suggest that adjacent tumor and PIN lesions shared genetic alterations, whereas distant lesions did not. Moreover, the observation of shared allelic losses at the NEFL locus in tumor and PIN lesions, but loss at D8S87 in tumors only, suggests that deletion of proximal and centromeric 8p sequences may comprise 645
1 2 3 4
B FIGURE 1. Allelic loss at the NEFL locus in prostatic intraepithelial neoplasia (PIN) and tumor. Examples of allelic loss at NEFL are shown for tumor (Tl), adjacent PIN (PI), distant PIN (P2), and normal (N) specimens from cases 7 (A) and 24 (B). The lower (3) and upper (2) alleles are deleted in Tl and P2, respectively, in case 7 (A). Note the presence of a weak instability for the upper allele in P2. The lower allele (4) is deleted in Tl and PI of case 24, whereas the upper allele (21 is deleted in P2 (B).
“early” and “late” events, respectively, tate tumorigensis.
PIN AND PCA SHAREA “GENETIC HISTORY” ALONG THE PATHWAY TO PROSTATETLJMORIGENESIS
The results of the studies presented here show that adjacent PIN and tumor foci share loss of the same allele at the NEFL locus more frequently than distant foci. Three interpretations may be applied to these findings: (1) Adjacent PIN and tumor lesions evolved from a common precursor lesion that suffered the initial genetic alteration. However, this interpretation is unlikely, as no candidate common precursor lesions (eg, adenomas) with such genetic alterations were defined in the prostate. (2) It is possible that both dysplastic and malignant cells experienced common but independent 8p deletion events during their evolution. This prospect cannot be ruled out by the studies performed here; however, it would require the coincidental occurrence of identical genetic alterations in PINS and carcinomas, events that are undocumented elsewhere. (3) PIN cells may experience specific genetic alterations that are permissive for, or facilitate, malignant transformation within an environment of genetic change. If so, such genetic alterations would be “passed on” to malignant cells, which could then develop into one or more tumor foci spatially adjacent to the original dysplasia. This interpretation is consistent with the hypothesis that dysplastic PIN lesions may evolve into malignant tumors in the prostate. 646
From these considerations, it appears that the results of the current study are primarily consistent with the hypothesis that PIN and tumor foci share a common “genetic history” (eg, allelic loss at the AJEFL locus) during malignant evolution in the prostate. It is unlikely, however, that examination of clinical samples alone will further delineate this process. Rather, future studies would benefit from focusing on the observation of prostate tumor progression in vivo, perhaps by utilizing more dynamic systems, such as the Transgenic Mouse Prostate (TRAMP)‘l or gene knockout animal models. In these models, the chronologic appearance of dysplastic lesions before malignant foci, or the observation of malignant glands “budding off’ from PIN lesions, would provide more direct evidence that PIN lesions evolve into malignant foci in the prostate. 8,22The use of such models to examine the relationship between PIN and carcinoma could greatly facilitate the elucidation of the biologic mechanisms underlying tumorigenesis in the prostate. REFERENCES 1. McNeal JE, and Bostwick DG: Intraductal dysplasia: a premalignant lesion of the prostate. Hum Path01 17: 64-71, 1986. 2. Bostwick DG: Prospective origins of prostate carcinoma. Cancer 78: 330-336,1996. 3. Bostwick DG: High grade prostatic intraepithelial neoplasia. Cancer Detect Prev Suppl 75: 1823-1836, 1995. 4. Ferguson J, Zincke H, Elhson E, Bergstrahl E, and Bostwick DG: Decrease of prostatic intraepithelial neoplasia following androgen deprivation therapy in patients with stage T3 carcinoma treated by radical prostatectomy. Urology 44: 91-95,1994. 5. Raviv G, Janssen T, Zlotta AR, Descamps F, Verhest A, and Schulman CS: Prostatic intraepithelial neoplasia: influence of clinical and pathological data on the detection of prostate cancer. J Urol 156: 1050-1054, 1996. 6. Davidson D, Bostwick DG, Qian J, Wollan PC, Oesterling JE, Rudders RA, Siroky M, and Stilmant M: Prostatic intraepithelial neoplasia is a risk factor for adenocarcinoma: predictive accuracy in needle biopsies. J Urol 154: 1295-1299, 1995. 7. Shepherd D, Keetch DW, Humphrey PA, Smith DS, and Stahl D: Repeat biopsy strategy in men with isolated prostatic intraepithelial neoplasia on prostate needle biopsy. J Urol 156: 460-463, 1996. 8. McNeal JE, Villers A, Redwine EA, Freiha FS, and Stamey TA: Microcarcinoma in the prostate: its association with duct-acinar dysplasia. Hum Path01 22: 644-652, 1991. 9. de la Torre M, Haggman M, Brandstedt S, and Busch C: Prostatic intraepithelial neoplasia and invasive carcinoma in total prostatectomy specimens: distribution, volumes and DNA ploidy. Br J Ural 72: 207-213, 1993. 10. Sakr WA, Haas GP, Cassin BF, Pontes JE, and Crissman JD: The frequency of carcinoma and intraepithelial neoplasia of the prostate in young males. J Urol 150: 379385, 1993. 11. Sakr WA, Grignon DJ, Crissman JD, Heilbrun LK, Cassin BJ, Pontes JE, and Haas GP: High grade prostatic intra-
UROLOGY 50 (4), 1997
epithelial neoplasia (HGPIN) and prostatic adenocarcinoma between the ages of 20 and 69: an autopsy of 249 cases. In Vivo 8: 439-443, 1994. 12. Sakr WA, Macoska JA, Benson P, Grignon DJ, Wolman SR, Pontes JE, and Crissman JD: Allelic loss in locally metastatic, multisampled prostate cancer. Cancer Res 54: 32733277, 1994. 13. Emmert-Buck MR, Vocke CD, Pozzatti RO, Duray PH, Jennings SB, Florence CD, Zhuang Z, Bostwick DG, Liotta LA, and Linehan WM: Allelic loss on chromosome 8~12-21 in microdissected prostatic intraepithelial neoplasia. Cancer Res 55: 2959-2962, 1995. 14. Qian J, Bostwick DG, Takahashi S, Bore11 TJ, Herath JF, Lieber MM, and Jenkins RB: Chromosomal anomalies in prostatic intraepithelial neoplasia and carcinoma detected by fluorescence in situ hybridization. Cancer Res 55: 54085414,1995. 15. Grignon DJ, and Sakr WA: Pathologic staging of prostate carcinoma: what are the issues? Cancer 78: 337-340, 1996. 16. Gleason DF: Histologic grading of prostate cancer: a perspective. Hum Path01 23: 273-279, 1992.
50 (41, 1997
17. Macoska JA, Benson PD, Turkeri LN, Haas GP, and Sakr W: PCR-based analysis of DNA from autopsied prostate tissue. PCR Meth Appl 2: 354-355, 1993. 18. Macoska JA, Trybus TM, Sakr WA, Wolf MC, Benson PD, Powell IJ, and Pontes JE: Fluorescence in situ hybridization analysis of 8p allelic loss and chromosome 8 instability in human prostate cancer. Cancer Res 54: 3824-3830, 1994. 19. Rogaev E, Rogaeva E, Lukiw WJ, Vaula G, Liang Y, Hancock R, McLachlan DC, and St George-Hyslop PH: An informative microsatellite repeat polymorphism in the human neurofilament light polypeptide gene (NEFL). Hum Mol Genet 1: 781, 1992. 20. Weber JL, Kwitek AE, May PE, Patterson D, and Drabkin H: Dinucleotide repeat polymorphisms at the D8S85, D8S87 and D8S88 loci. Nucl Acids Res 18: 4038, 1990. 21. Greenberg NM, DeMayo F, Finegold MJ, Medina D, Tilley WD, Aspinall JO, Cunha GR, Donjacour AA, Matusik RJ, and Rosen JM: Prostate cancer in a transgenic mouse. Proc Nat1 Acad Sci USA 92: 3439-3443,1995. 22. Perez-Stable C, Altman NH, Mehta PP, Deftos LJ, and Roos BA: Prostate cancer progression, metastasis and gene expression in transgenic mice. Cancer Res 57: 900-906, 1997.