p53 protein overexpression in smooth muscle tumors of the uterus

p53 Protein Overexpression in Smooth Muscle Tumors of the Uterus THEODORE H. NIEMANN, MD, STEPHEN S. RAAB, MD, JULIA C. LENEL, PHD, JANIS R. RODGERS, BS, AND ROBERT A. ROBINSON, MD, PHD p53 is a nuclear phosphoprotein whose overexpression may portend a poor prognosis in a variety of neoplasms. In this immunohistochemical study we examined p53 overexpression in a variety of uterine smooth muscle tumors (34 leiomyosarcomas, 18 leiomyomas, and six smooth musde tumors of uncertain malignant potential [STUMPs]). p53 immunoreactivity was observed in none of 18 (0%) leiomyomas, one of six (17%) STUMPs, and 16 of 34 (47%) leiomyosarcomas. Reactivity was not observed in the surrounding nouneoplastic uterine smooth muscle. Strong p53 overexpression in the leiomyosarcomas was significantly associated with high grade morphology (P = .013) and a high stage at the time of presentation (P = .021). In

25 leiomyosarcoma patients with clinical follow-up, p53 overexpression was associated with shorter length of survival (P = 0.024). However, this effect was not independent of tumor stage or grade. A regression analysis showed that tumor stage was the only independent predictor of length of survival. Our study size is small, and further studies are warranted to determine the significance and replicability of these findings. HUM PATHOL 26:375--379. Copyright © 1995 by W.B. Saunders Company Key words: p53, leiomyoma, leiomyosarcoma, prognosis. Abbreviations: STUMP, smooth muscle tumors of uncertain malignant potential; HE, hematoxylin-eosin.

Smooth muscle tumors are the most common neop l a s m s o f t h e uterus. T h e vast m a j o r i t y o f t h e s e a r e l e i o m y o m a s , w h i c h a r e e s t i m a t e d to o c c u r in 20% to 30% o f w o m e n o l d e r t h a n 30. I n c o m p a r i s o n , l e i o m y o s a r c o m a s a r e relatively rare, r e p r e s e n t i n g a p p r o x i m a t e l y o n e o f every 800 u t e r i n e s m o o t h m u s c l e t u m o r s a n d a c c o u n t i n g f o r 1.8% o f u t e r i n e c o r p u s m a l i g n a n cies. 1'2 I n m o s t cases, t h e h i s t o l o g i c a l d i s t i n c t i o n between a l e i o m y o m a a n d a l e i o m y o s a r c o m a is n o t a diagn o s t i c d i l e m m a . H o w e v e r , in a small p e r c e n t a g e o f t u m o r s this d i f f e r e n t i a t i o n m a y b e difficult. I n d i v i d u a l h i s t o l o g i c a l f e a t u r e s , s u c h as m i t o t i c activity o r cytologic atypia, l a c k s p e c i f i c i t y J '4 I n s t e a d , a n a l y z i n g a c o m b i n a t i o n o f m u l t i p l e f e a t u r e s s e e m s to b e t h e m o s t effective a p p r o a c h to d i a g n o s i n g t h e s e p r o b l e m a t i c t u m o r s . 5 T h i s d i s t i n c t i o n is i m p o r t a n t b e c a u s e l e i o m y o s a r c o m a s a r e aggressive n e o p l a s m s t h a t r e s u l t in a p o o r p r o g n o sis. 2 p 5 3 is a n u c l e a r p h o s p h o p r o t e i n t h a t c a n r e g u l a t e cell p r o l i f e r a t i o n a n d s u p p r e s s t u m o r growth. 6 Mutations in t h e p53 g e n e have b e e n r e p o r t e d in a variety o f h u m a n t u m o r s , a n d in s e l e c t e d m a l i g n a n c i e s overexp r e s s i o n o f p53 has b e e n a s s o c i a t e d with a p o o r p r o g n o sis. 7-11 I n this s t u d y we e x a m i n e d a series o f u t e r i n e s m o o t h m u s c l e t u m o r s to d e t e r m i n e if p53 is overexp r e s s e d in t h e s e n e o p l a s m s a n d if o v e r e x p r e s s i o n can b e u s e d to p r e d i c t t h e b e h a v i o r o f these t u m o r s .

retrospectively for cases of uterine leiomyosarcoma. Fortythree cases were identified. In six of these cases paraffin blocks were not available. Upon review three of the remaining cases were downgraded (two to smooth muscle tumor of uncertain malignant potential [STUMP] and one to leiomyoma). This left 34 cases of uterine leiomyosarcoma with material available for analysis. In addition, six uterine STUMPs and 18 uterine leiomyomas were selected for analysis. The original pathological material was reviewed and assessed for cellularity, mitoses, necrosis, and infiltrating borders. On the basis of these findings the tumors were classified as leiomyoma, STUMP, or leiomyosarcoma. 5'~I The leiomyosarcomas were then divided into low, intermediate, and high grade. Clinical history and follow-up data were obtained through review of the patients' charts. This information was available for 47 of the 58 patients (18 of 18 leiomyomas, four of six STUMPs, 25 of 34 leiomyosarcomas). Specific features recorded included age, menstrual status, stage of disease at the time of diagnosis, 13 adjuvant therapy, time to recurrence or metastasis, site of metastasis, and overall length of survival. Immunohistochemical staining was performed on formalin-fixed, paraffin-embedded tissue sections. The sections were cut 3-#m thick, deparaffinized in xylene, and rehydrated with graded alcohols. The slides underwent antigen retrieval using 1% zinc sulfate. Endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide in methanol. Two antibodies to p53 were used on separate slides from the same tumor. D07 (Dako, Carpinteria, CA) is a murine monoclonal antibody that recognizes both wild and mutant p53. CM1 (Signet, Dedham, MA) is a rabbit polyclonal antibody that also recognizes both wild and mutant p53. The antibodies were applied at a dilution of 1:50 and allowed to incubate for 2.5 hours at room temperature. The signal was detected using a commercially available linked streptavidin-biotin-peroxidase system (Dako) with diaminobenzidine as the chromogen and 10% Harris' hematoxylin as a counterstain. A section from a colon carcinoma with documented p53 overexpression was used as an external positive control. Negative controls consisted of a complimentary section from each tumor with substitution of nonimmune serum for the primary antibody. The specimens were assessed for the presence of nuclear reactivity. An absence of nuclear reactivity was considered nonreactive. Three specimens demonstrated a fine stippling

MATERIALS AND METHODS The files of the Department of Pathology at the University of Iowa Hospitals and Clinics, Iowa City, IA, were searched From the Department of Pathology, University of Iowa, Iowa City, IA. Accepted for publication August 15, 1994. Presented in part at the United States-Canadian Academy of Pathology Meeting, San Francisco, CA, March 12 to 18, 1994. Address correspondence and reprint requests to Theodore H. Niemann, MD, Department of Pathology, The Ohio State University Hospital, 403 E Doan Hall, 410 W 10th Ave, Columbus, OH 43210. Copyright © 1995 by W.B. Saunders Company 0046-8177/95/2604-000355.00/0

375

HUMAN PATHOLOGY

Volume 26, No. 4 (April 1995)

FIGURE 1. (A) Leiomyoma showing fasicles of uniform spindled cells. (B) Immunohistochemical staining for p53(D07) shows an absence of nuclear reactivity. (A, hematoxylin-eosin (HE); B, anti-p53 (D07); original magnification ×400.)

of a small percentage of the nuclei (<5%) with one antibody. This pattern was identified only on high power examination and was considered plus/minus reactivity. Dark staining of the entire nucleus in numerous cells (20% to 60%) was considered 1 + reactivity and was able to be visualized at low magnification. A chi-squared test was used to determine the association between p53 overexpression, tumor grade, tumor stage, and follow-up status (alive or dead). The Kaplan-Meier product limit method for censored data was used to plot separate survival functions for reactive and nonreactive leiornyosarcomas and to analyze the relationship between p53 overexpression and lengths of survival. Additional survival functions were plotted after stratifying the sample on tumor grade and stage. Cox's log-linear regression model was used to determine the impact of p53 overexpression, tumor grade, and stage on length of survival)4

RESULTS p53 immunoreactivity was observed in n o n e of 18 (0%) leiomyomas (Fig 1), one of six (17%) STUMPs, and 16 of 34 (47%) leiomyosarcomas. Reactivity was typically strong and diffusely distributed t h r o u g h o u t the t u m o r (1+) (Fig 2). In general, the pattern of nuclear reactivity was comparable between the two antibodies. However, two low grade leiomyosarcornas and the one p53-reactive STUMP showed focal, weak reactivity

376

(plus/minus) with only the D07 antibody. T h e r e was no reactivity in the adjacent nonneoplastic myometrium. For statistical evaluation the specimens showing weak reactivity (plus/minus) were included with the nonreactive specimens. When leiomyosarcomas were divided by grade, the percentage of reactive specimens increased with grade. N o n e of five (0%) low grade, three of 12 (25%) intermediate grade, and 11 of 17 (65%) high grade leiomyosarcomas showed strong reactivity (X 2 = 9.63; df = 3; P = .021) (Table 1). W h e n the leiomyosarcomas were analyzed by stage, six of 16 stage 1, n o n e of two stage 2, six of seven stage 3, and two of three stage 4 tumors showed strong p53 reactivity (X z = 9.693; df = 3; P = 0.21) (Table 2). T h e 18 patients with leiomyomas are alive and well. T h e four patients with STUMPs and follow-up also are alive and well. This includes the patient with the STUMP showing weak p53 reactivity. O f the 25 patients with leiomyosarcoma and follow-up, 16 of 25 (64%) are dead of disease (range, 2 to 83 months; mean, 19 months; median, 12 months), five are alive with disease (range, 9 to 45 months; mean, 31 months; median, 33 months) and four are alive and well (range, 12 to 61 months; mean, 41.5 months). A chi-squared test of association showed no significant association between p53 overexpression and ultimate o u t c o m e (dead v alive)

p53 IN UTERINESMOOTH MUSCLE TUMORS (Niemann et al)

FIGURE 2. (A) Leiomyosarcoma showing fasicles of cells with irregular, hyperchromatic nuclei and frequent mitotic figures. (B) Immunohistochemical staining for p53 (D07) shows reactivity in a majority of the tumor nuclei. (A, HE; B, anti-p53 (D07); original magnification ×400.)

grade, and p53 overexpression were entered into Cox's log-linear regression model, only stage was significantly associated with length of survival (P = .005).

(P : .282). However, the survival functions for nonreactive and reactive specimens were not equal (generalized Savage Wilcoxin's test = 5.127; df= 1; P = .024) with shorter length of survival associated with reactive specimens (Fig 3). To determine whether the relationship was i n d e p e n d e n t of stage and grade of tumor, stratified survival functions were computed. Because of the sparsity of the data, both grade and stage were collapsed into two levels (low and intermediate v high grade, stages 1 and 2 v 3 and 4). When either grade or stage was used to stratify the sample, p53 overexpression and length of survival were no longer significantly related (P = .500 and P = .320, respectively). When stage,

TABLE 1.

DISCUSSION p53 is a 53-kd nuclear phosphoprotein that in its wild form functions to control cellular proliferation and suppress tumor growth. 6 The mechanism of this activity remains speculative. 15 Mutations in the p53 gene lead to a loss of normal regulatory function and have been

TABLE 2.

Relationship Between p53 Expression and Grade

Relationship Between p53 Expression and Stage p53

p53 Grade

Nonreactive

Reactive

Total

1 2 3 TotN

5 9 6 20

0 3 11 14

5 12 17 34

N O T E . X z = 8.682;

df =

2; P = 0.013.

Stage

Nonreactive

Reactive

Total

1 2 3 4 Total

16 2 1 1 20

6 0 6 2 14

22 2 7 3 34

N O T E . X 2 = 9.693;

377

df=

3; P = 0.021.

HUMAN PATHOLOGY

Volume 26, No. 4 (April 1995)

-7 I I

I

0.8 I I

Nonreactive

I I ~-]

0.6

I I

Proportion Surviving

I I

0.4 Reactive

0.2

0 0

i I i I i I r I ~ I i I p I i I -~-10 20 30 40 50 60 70 80 90 Survival Time (Months)

FIGURE 3. Kaplan-Meier survival curve showing that p53 overexpression is associated with a shorter length of survival (P = 0.024). The median length of survival for patients with p53 overexpression is 7.0 months c o m p a r e d with 31.8 months for patients without p53 overexpression.

observed in a wide variety of h u m a n neoplasms. 7,s These mutations frequently lead to the production of an altered protein with a prolonged half-life. As this protein accumulates it becomes detectable by standard immunohistochemistry, a6T h e r e is not an absolute correlation between p53 overexpression as detected by immunohistochemistry and p53 gene mutation as detected by molecular techniques. On the one hand, p53 overexpression has been reported in the absence of d o c u m e n t e d genetic alterations. 16'a7 On the other hand, mutations may lead to loss of p53 production or significant alteration in the protein with a resulting absence of immunohistochemical reactivity, as Nonetheless, immunohistochemistry provides a relatively simple technique to initially study p53 alterations. p53 has been examined in a n u m b e r of uterine neoplasms with several studies reporting p53 overexpression in uterine adenocarcinomas. Using frozen tissue and monoclonal antibody Pab1801, Bur et a119 reported p53 overexpression in 59% of endometrial adenocarcinomas. They noted that p53 overexpression was observed more frequently in uterine serous carcin o m a compared with typical endometrioid adenocarcinoma. They also reported that p53 overexpression significantly correlated with high surgical stage. Using 378

frozen tissue and monoclonal antibody Pab1801, Kohler et al 2° reported that p53 overexpression occurred in 21% of endometrial adenocarcinomas. They also reported a significant association of p53 overexpression with n o n e n d o m e t r i o i d histology and high surgical stage. p53 overexpression has been less extensively studied in uterine mesenchymal neoplasms. Bur et a119 included a small n u m b e r of mesenchymal tumors in their study of uterine neoplasms. They reported p53 overexpression in two of four malignant mixed mflllerian tumors, but reactivity was not observed in two endometrial stromal sarcomas, two leiomyomas, and a single leiomyosarcoma. Liu et a121 examined a larger series of mesenchymal neoplasms and reported p53 overexpression in 26 of 41 mixed mesodermal tumors, one of four leiornyosarcomas, and none of one endometrial stromal sarcoma. In this study we examined p53 overexpression in a series of uterine smooth muscle tumors. By immunohistochemistry strong p53 reactivity was not detected in the smooth muscle tumors of uncertain malignant potential or the leiomyomas. In contrast, strong p53 reactivity was noted in 14 of 34 (41%) leiomyosarcomas. This indicates that p53 overexpression in smooth muscle tumors is relatively specific for leiomyosarcoma and potentially could be used as a diagnostic marker of a malignant phenotype. However, there is a significant association between p53 reactivity and high grade morphology. This indicates that most of the tumors that overexpressed p53 were high grade and did not represent diagnostic dilemmas. The prognostic use of p53 overexpression seems to vary with the tumor site. Studies have suggested that p53 overexpression in breast carcinoma, gastric carcinoma, and prostatic carcinoma is associated with a significantly shortened length of survival, 9-H whereas studies of colorectal carcinoma and cervical carcinoma have failed to demonstrate an i n d e p e n d e n t prognostic significance of p53 overexpression. 22'2~ In this study we examined the relationship between p53 overexpression in leiomyosarcomas and length of survival. Although p53 overexpression is associated with a shorter length of survival, this effect is not i n d e p e n d e n t of stage and grade of disease. Using Cox's log-linear regression model, stage was the only factor significantly associated with length of survival (P = .005). In conclusion, p53 overexpression was frequently observed in uterine leiomyosarcomas and was associated with high tumor grade, high stage at presentation, and shorter length of survival. Uterine leiomyosarcomas are relatively u n c o m m o n neoplasms. Consequently, our study size is small and our data analyses may lack power. Additional studies are necessary to determine the genetic basis for this overexpression and to further clarify the potential diagnostic, prognostic, and therapeutic use of these findings.

Acknowledgment. We gratefully acknowledge the secretarial assistance of Katie Dasse and the photographic assistance of Joel Carl.

p53 IN UTERINESMOOTH MUSCLE TUMORS (Niemann et al)

REFERENCES 1. Zaloudek C, Norris HJ: Mesenchymal tumors of the uterus, in Kurman RJ (ed) : Blaustein's Pathology of the Female Genital Tract (ed 3). New York, NY, Springer-Verlag, 1987, pp 373408 2. Platz CE, BendaJA: Female genital tract cancer incidence and prognosis by histologic type, SEER population based data 1973-1987. Cancer (in press) 3. Perrone T, Dehner LP: Prognostically favorable "mitotically active" smooth-muscle tumors of the uterus: A clinicopathologic study of ten cases. A m J Surg Pathol 12:1-8, 1988 4. Prayson RA, Hart WR: Mitotically active leiomyomas of the uterus. A m J Clin Pathol 97:14-20, 1992 5. Bell SW, Kempson RL, Hendrickson MR: Problematic uterine smooth muscle neoplasms: A clinicopathologic study of 213 cases. Am J Surg Pathol 18:535-558, 1994 6. Vogelstein B, Kinzler KW: p53 Function and dysfunction. Cell 70:523-526, 1992 7. NigroJM, Baker SJ, Preisinger AC, et al: Mutations in the p53 gene occur in diverse human tumour types. Nature 342:705-708, 1989 8. Porter PL, Gown AM, Kramp SG, et al: Widespread p53 overexpression in human malignant tumors: An immunohistochemical study using metacarn-fixed, embedded tissue. Am J Pathol 140:145153, 1992 9. Thor AD, Moore DH, Edgerton SM, et al: Accumulation of p53 tumor suppressor gene protein: An independent marker of prognosis in breast cancers, J Natl Cancer Inst 84:845-855, 1992 10. Martin HM, Filipe MI, Morris RW, et ah p53 Expression and prognosis in gastric carcinoma. I n t J Cancer 50:859-862, 1992 11. Visakorpi T, Kallioniemi OP, Heikkinen A, et al: Small subgroup of aggressive, highly proliferative prostatic carcinomas defined by p53 accumulation. J Natl Cancer Inst 84:883-887, 1992

379

12. Silverberg SG, Kurman RJ: Smooth muscle and other mesenchymal tumors, in Rosai J (ed): Tumors of the Uterine Corpus and Gestational Trophoblastic Disease. Atlas of Tumor Pathology, third series. Washington, DC, Armed Forces Institute of Pathology, 1992, pp 113-152 13. Creasman WT: New gynecologic cancer staging. Obstet Gynecol 75:287-288, 1990 14. Cox DR: Regression models and life tables. J R Stat Soc 34:187-220, 1972 15. PietenpolJA, Vogelstein B: No room at the p53 inn. Nature 365:1%18, 1993 16. Hollstein M, Sidransky D, Vogelstein B, et al: p53 Mutations in human cancers. Science 253:49-53, 1991 17. Kohler MF, Marks JR, Wiseman W, et al: Spectrum of mutation and frequency of allelic deletion of the p53 gene in ovarian cancer. J Natl Cancer lust 85:1513-1519, 1991 18. Risinger JI, Dent GA, Ignar-Trowbridge D, et al: p53 Gene mutations in human endometxial carcinoma. Mol Carcinog 5:250253, 1992 19. Bur ME, Perlman C, Edelmann L, et ah p53 Expression in neoplasms of the uterine corpus. A m J Clin Pathol 98:81-87, 1992 20. Kohler MF, Berchuck A, DavidoffAM, et al: Overexpression and mutation of p53 in endometrial carcinoma. Cancer Res 52:16221627, 1992 21. Liu F-S, Kohler MF, Marks JR, et al: Mutation and overexpression of p53 tumor suppressor gene frequently occurs in uterine and ovarian sarcomas. Obstet Gynecol 83:118-124, 1994 22. Scott N, Sagar P, Stewart J, et al: p53 In colorectal cancer: Clinicopathological correlation and prognostic significance. BrJ Cancer 63:317-319, 1991 23. Oka K, Nakano T, Arai T: p53 CM1 expression is not associated with prognosis in uterine cervical carcinoma. Cancer 72:160163, 1993