國家圖書館典藏電子全文

Size: px

Start display at page:

Download "國家圖書館典藏電子全文"

顺僻飘终
7 years ago
Views:

1 Master Thesis, Institute of Medicine, Chung-Shan Medical and Dental College MULTIVARIATE ANALYSIS OF THE HISTOPATHOLOGIC PROGNOSTIC FACTORS FOR CERVICAL CANCER PATIENTS TREATED BY SURGERY AND POST- OPERATIVE RADIATION Advisor: Ming-Chih Chou, M.D., Ph.D. Graduate Student: Kun-Goung Lai, M.D.

FOR CERVICAL CANCER PATIENTS TREATED BY SURGERY AND POST- OPERATIVE

2 ABSTRACT Background : Cervical cancer is by now still keeping its high rank incidence as one of the most popular cancers of women in Taiwan. Although its mortality rate is modest if treated properly, however, some early stage cancers of the uterine cervix still behave out of our expectation despite surgery and post-operative irradiation. Probably several factors exist which indeed affect the ultimate prognosis of early stage cervical cancer other than clinical staging. We have investigated detailed pathological reports from hysterectomy specimen and evaluate whether some histopathological factors might really play major roles in predicting the prognosis of early stage carcinoma of the uterine cervix. Materials and Methods: Between 1st, Dec and 31, Dec. 1991, 126 patients with pathology-proven stages I and II carcinoma of the uterine cervix were treated by radical hysterectomy and pelvic lymphadenectomy ( some of them undergone ATH ) followed with post-operative irradiation at Department of Radiation Oncology, Chang-hua Christian Hospital. The indications for radiotherapy were based on pathologic findings including lymph node metastasis, positive surgical margins, lymphvascular permeation, primary tumor size 3.0 cm, parametrial involvement, corpus invasion, previous inadequate surgery ( ATH ) or more than one combination of the above. Patients usually received external beam irradiation 39.6 Gy to 61.2 Gy to the whole pelvis, and the vaginal cuff were boosted by intravaginal brachytherapy. All patients who received complete course of radiotherapy had minimal follow-up period of at least nine years ( 108 months). The prognosis-related histologic parameters including histologic subtypes, tumor grade, longitudinal tumor dimension, lymph-vascular space invasion, corpus invasion, parametrial extension, vaginal invasion and pelvic lymph node ( PLN ) IX

Probably several factors exist which indeed affect the ultimate prognosis of early stage cervical cancer other than clinical staging.

3 metastases were collected and evaluated with univariate and multivariate analysis. Results : All the 126 cases were evaluated, analyzed and followed until closure of this study( Jun. 2001). Half ( 63 ) of all the cases died. The median follow-up duration was 113 months ( range: ). The survival duration was calculated from time of histologic diagnosis to the time of last follow-up or death. The estimated 5- and 10-year survival rate of surgical stages IB, IIA and IIB were 65%, 75%, 62% and 59%, 49%, 53% respectively ( p > 0.05). Univariate analysis revealed that PLN metastases, histologic subtype, tumor size and corpus invasion were significant in survival ( p <0.05 ). Among these variables, we found that histologic subtype and PLN metastases were identified as the most independent and significant prognostic factors ( p <0.005) by multivariate analysis using Cox regression model. Discussions : In the current study, we found that histologic subtype and PLN metastases were recognized as two independent and significant prognostic variables by Cox s model. For number of metastatic nodes more than two and non-scc from pathologic classification ( especially adenocarcinoma ), the prognosis might be the worst. This retrograde analytical survey of various postoperative histopathological prognostic factors indicates a need for subdivision by more significant risk factors which could predict the outcome of surgically resectable early-stage cervical cancer more precisely than clinical staging and also serves as a new indicator for changing treatment modality. Key words : Uterine cervix, Prognostic factors, Multivariate analysis X

The survival duration was calculated from time of histologic diagnosis to the time of last follow-up or death.

4 .,,.,. ( ),. : 126 ( ).,,, 3.0,, Gy,. VI

5 ( 108 ).,,,,,,. : 126, ( ). ( 63 ), 113 ( : ).., IB, IIA, IIB 65%, 75%, 62% 59%, 49%, 53% ( p >0.05).,,, ( p <0.05). Cox regression model, ( p <0.005 ). : Cox s model. ( VII

6 ),.,,,. VIII

7 I., ( International Federa- tion of Gynecology and Obstetrics, FIGO ).,, ( high-risk group).,., 1,4,5,7,19,20,22, 1,3,4, 18,19,20,25, 1,3,6,19,25, 1,3,6,18, 1,22,24, 1,5, 19,25 1,25,.,. I (IB) II (A, B),. 126, 1

8 ,. 2

9 **** **** II--V VI--VIII IX--X Kaplan-Meier I

-------------------------------------- 4 3. ----------------------------- 5 ------------------------------------------ 7 1.

10 II ,500 cgy, 1, 3, , 76 ( 51.72, 52, 10.13). ( Fig. 1), IB 57 ( 45.2 %), IIA 28 ( 22.2 %), IIB 41 ( 32.6 %)., 111 ( 88.1 %, 24, 42, 3, 42 ), 7 ( 5.5 %), 8 ( 6.4 %).,,,, ;. 3

11 ., 102 ( 81 %), 24 ( 19 %)., :, 3,,,,,, ( 22 )., cm, ( ), cGy,. 4,400 5,040 cgy,, 1,080 1,400 cgy., ( Cs-137 ),, cm 500 cgy., 8 4

12 , 256, SPSS. Kaplan-Meier, Mantel-Cox test ( Log-rank test ). ( logistic regression analysis ). 5 % Cox s proportional-hazard s regression model. forward stepwise procedure. 5

13 Fig. 1 Age Distribution of Cervical Cancer, S/P RH or ATH 6

14 III. 1. Kaplan-Meier ( ), ( : 8, 256 ). Kaplan-Meier survival analysis IB, IIA IIB 65%, 75%, 62% ; 59%, 49% 53% ( p =.8531, Fig. 2) Fig. 2 Kaplan-Meier survival curve by surgical staging ( 5: IB, 2: IIA, 6: IIB, p =.8531 ) 7

15 2. Mantel-Cox test, ( Fig Tab. 1). Fig. 3 Kaplan-Meier survial curve by patient s age ( 1: 45 yr, 2: yr, 3: 55 yr, p =.1875 ) 8

16 Fig. 4 Kaplan-Meier survival curve by histology ( a: Squamous cell carcinoma, b: Adenocarcinoma & adenosquamous carcinoma, p =.0112 ) 9

17 Fig. 5 Kaplan-Meier survival curve by tumor diameter ( 1: tumor size 3 cm, 2: tumor size 3 cm, p =.0303 ) 10

18 Fig. 6 Kaplan-Meier survival curve by presence or absence of pelvic lymph node metastasis ( 0: No metastasis, 1: Positive for metastasis, p =.1199 ) 11

19 Fig. 7 Kaplan-Meier survival curve by number of pelvic lymph node(s) metastases ( 0 = Negative, 1 = No. 1-2, 2 = No. > 2, p =.0016 ) 12

20 Fig. 8 Kaplan-Meier survival curve by presence or absence of lymph-vascular space invasion, LVSI ( 0 = Negative, 1 = Positive, p =.5416) 13

21 Fig. 9 Kaplan-Meier survival curve by presence or absence of parametrial invasion ( 0 = Negative, 1 = Positive, p =.4225 ) 14

22 Fig. 10 Kaplan-Meier survival curve by presence or absence of corpus invasion ( 0 = Negative, 1 = Positive, p =.0371 ) 15

23 Fig. 11 Kaplan-Meier survival curve by presence or absence of vaginal invasion ( 0 = Negative, 1 = Positive, p =.8566 ) 16

24 Fig. 12 Kaplan-Meier survival curve by mode of treatment ( 1 =S( Rec) + R/T, 2 = S + R/T and S + C/T + R/T, p =.0082 ) 17

25 Tab. 1 Survival Analysis for Prognostic Factors using Mantel-Cox Test No. of Estimated survival rate(%) Variables Patients 5-yr 10-yr p-value PLN mets 100 None > Histol. subtype 126 SCC Non-SCC Tumor diameter cm cm Lymph-vascular 104 space invasion No Yes N.S. Param. invasion 115 No Yes N.S. Corpus invasion 118 No Yes Vaginal invasion 114 No Yes N.S. Treatment mode 122 S C/T+R/T S(Rec)+R/T *PLN: pelvic lymph node, SCC: squamous cell carcinoma, Rec: recurrence 18

26 ,, ( p=0.1199)., (1 0) ( p= ). ; ( p=0.0112). 3 3 ( p= )., ( ) ( ) ( ),. Mantel- Cox test, ( ) Cox s proportional hazards model 15,16,17. forward stepwise procedure ( Tab. 2 ). 19

27 Tab.2 Cox Regression Analysis with Covariates Categorical Variable Codings a,b Frequency (1) HISTOLOGY a b a. Indicator Parameter Coding b. Category variable: HISTO_3 1. Block 0: Beginning Block Variables not in the Equation a,b Score Df Sig. PLN_METS HISTOLOGY T_DIAM CORPUS INV a. Residual Chi Square = with 4 df Sig. =0.000 b Beginning Block Number 0, initial Log Likelihood function: -2 Log likelihood: Block 1: Method = Forward Stepwise (Wald) Omnibus Tests of Model Coefficients c,d Step -2 Log Overall (score) Change From Previous Step Change From Previous Block Likelihood Chi-square Df Sig. Chi-square df Sig. Chi-square df Sig. 1 a b a. Variable(s) Entered at Step Number 1:HISTOLOGY b. Variable(s) Entered at Step Number 2:PLN_METS c. Beginning Block Number 0, initial Log Likelihood function: -2 Log likelihood: d. Beginning Block Number 1, Method: Forward Stepwise (Wald) 20

28 Variables in the Equation B SE Wald df Sig. Exp(B) Step1 HISTOLOGY Step2 PLN_METS HISTOLOGY Variables not in the Equation a,b Score Df Sig. Step 1 PLN_METS T_DIAM CORPUS Step 2 T_DIAM CORPUS a. Residual Chi Square = with 3 df Sig. = b. Residual Chi Square = with 2 df Sig. = Covariate Means Mean PLN_METS HISTOLOGY T_DIAM CORPUS

29 IV. Cox, Block 0 model, model,, d.f. = 1, model ( p =0.000 ). Block 1 forward stepwise method likelihood ratio. 2b, Log-Likelihood , ( d.f. =2, p=0.000 )., ( p >0.05 ). Wald test : Wald = N.D. (b/se(b)) b, SE(b) b. Wald, Cox model., Wald ( ) (14.113). ( p= ) Cox regression model 22

30 ( censored data),. 16,., ( ) ( p=0.0112), 1,2,22,24.,,., 1,22,24.? 111, 42 (37.8 %),., ( DFI, disease-free interval), 2,3,4,19.,., 23

31 ,., p 0.12,.,, 2 ( cut-off point ); : 2, 2 2 ( p=0.0016). ( Fig. 6-7 & Tab. 1), Kamura 1, PLN 0 1 ( p < 0.001). Alvarez 4 UAB 401, (pathologic dimensions) ( p ) ( p =0.0005) IB. Morita et al.( ) Tinga DJ 5,7, , Terry Fox Intern. Cancer Symposium., 24

32 ,.,. 4,.,,,. ( p= from log-rank test). Cox s regression model,.,,., , 23, ( erosion), ( friable), ( ill-defined) ( necrotic or slough ).,,. 1,2,4, giant frontal section stromal 25

33 invasion, ; 1, ( tumor size ),. 26

34 V.,, I, II ( ).. microvessel density 12,21,25, Human papillomavirus deoxyribonucleic acid 11, DNA ploidy & S-phase fraction 9,25 bcl-2 expression 14 Biomarker 25,. flow cytometry DNA, DNA index 1.5( 1.3 ),, 24,26, ; 27,?,? 27

35 ( )?, 3.0 2,,.,,. 28

36 VI. 1. Kamura T, Tsukamoto N, Tsuruchi N, et al. Multivariate analysis of the histopathologic prognostic factors of cervical cancer in patients undergoing radical hysterectomy. Cancer 1992; 69: Smiley LM, Burke TW, Silva EG et al. Prognostic factors in stage IB squamous cervical cancer patients with low risk for recurrence. Obstet Gynecol 77: 271, Delgado G, Bundy B, Zaino R, et al. Prospective surgical-pathological study of disease-free interval in patients with stage IB squamous cell carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 38, , Alvarez RD, Potter ME, Soong SJ, Gay FL et al. Rationale for using pathologic tumor dimensions and nodal status to subclassify surgically treated stage IB cervical cancer patients. Gynecol Oncol 43, , Inoue T and Morita K. The prognostic significance of number of positive nodes in cervical cacinoma stages IB, IIA and IIB. Cancer 65: , Zaino RJ, Ward S, Delgado MD, Bundy B, et al. Histopathologic Predictors of the behavior of surgically treated stage IB squamous cell carcinoma of the cervix ( A Gynecologic Oncology Group Study). Cancer 1992; 69: Tinga DJ, Timmer PR, et al: Prognostic significance of single versus multiple lymph node metastases in cervical carcinoma stage IB. Gynecol Oncol 39, , Lai KG et al: Cancer Registry Annual Report, Chang-hua Christian Hospital, Jelen I, Valente PT, et al: Deoxyribonucleic acid ploidy and S- phase fraction are not significant prognostic factors for patients with cervical cancer. Am.J. Obst. & Gynecol. 171(6): ,

37 10. Perez CA, Grigsby PW, Camel HM, et al. Irradiation alone or combined with surgery in stage IB, IIA and IIB carcinoma of uterine cervix: Update of a nonrandomized comparison. Int.J. Radiat. Oncol. Biol. Phys. Vol. 31, No. 4: , Rose BR, Thompson CH, Simpson JM et al. Human Papillomavirus deoxyribonucleic acid as a prognostic indicator in early-stage cervical cancer: A possible role for type 18. Am.J. Obst. & Gynecol. 173(5): , Bremer GL, Tiebosch A, van der Putten et al. Tumor angiogenesis: An independent prognostic parameter in cervical cancer. Am.J. Obst. & Gynecol. 174(1): , Jan Lai KG, Hsu RC and Ma YL. Definitive radiotherapy and postoperative irradiation in stages I, II carcinoma of uterine cervix: A six-year experience. Proc. of annual meetings of the Chinese Radiological Society, Mar Tjalma W, Weyler J, Goovaerts G. et al. Prognostic value of bcl-2 expression in patients with operable carcinoma of the uterine cervix. J. Clin. Pathol. 50(1): 33-36, Jan Christensen E, Schlichting P, Andersen PK, et al. Updating prognosis and therapeutic effect evaluation in cirrhosis with Cox s multiple regression model for time-dependent variables. Scand J Gastroenterol 1986, 21, Christensen E: Multivariate survival analysis using Cox s regression model. Hepatology Vol.7, No. 6, pp , Dawson-Saunders B, Trapp RG: Predicting a censored outcome --- proportional hazards model. Basic & Clinical Biostatistics, 2 nd edi. pp Monaghan JM: Prognostic factors in early cancer of the cervix with paticular reference to patients with negative lymph nodes. Proc. of the 1 st Annual Terry Fox & Chang Gung Memorial Hospital International Cancer Symposium on Cervical Cancer, Lai CH, Tang SG, Hsueh S, Hong JH, et al. Preoperative prognostic variables and impacts of postoperative adjuvant therapy in stage IB or II cervical carcinoma with or without pelvic nodes metastases : An analysis of 891 cases. Proc. of the 1 st Annual Terry Fox & CGMH International Cancer Symposium on Cervical Cancer,

38 20. Tsai CS, Hong JH, Chang JT, et al. The prognostic factors for early cervical cancer treated by radical hysterectomy and postoperative radiation. Proc. of the 1 st Annual Terry Fox & CGMH International Cancer Symposium on cervical Cancer, Obermair A, Wanner C, Bilgi S, et al. Tumor angiogenesis in stage IB cervical cancer: Correlation of microvessel density with survival. Am.J. Obst. & Gynecol. 178(2): , Feb Cohn DE, Peters WA III, Muntz HG: Adenocarcinoma of the uterine cervix metastatic to lymph nodes. Am.J. Obst. & Gynecol. 178(6): , Jun Brewster WR, DiSaia PJ, Monk BJ, et al. Young age as a prognostic factor in cervical cancer: Results of a population-based study. Am. J. Obst. & Gynecol. 180(6): , Jun Lai CH: Prognostic factors and role of chemotherapy in cervical cancer. Special lecture, the 5 th Taiwan Joint Cancer Conference, Perez CA, Brady LW: Principles and practice of radiation oncology, 3 rd edi. Philadelphia, Lippincott-Raven Publishers 1998; Jakobsen A, Bichel P, et al. Is radical hysterectomy always necessary in early cervical cancer? Gynecol Oncol 39, 80-81, Chen HJ, Lin FJ, Chen MS, et al. Combined postoperative pelvic irradiation in carcinoma of uterine cervix. Chin J Radiol 1992; 17: Fletcher RH, Fletcher SW, Wagner EH: Clinical Epidemiology The Essentials, 3 rd edi. Williams & Wilkins Asia-Pacific Ltd

39 ( )

40 ,, : :

41 ( ). 89 : : ( ),,,,,,, : :

42 ,,,.,,,!,,,,. 86, 87.,,,,,,,.,, ;,., II

43 ,, :,,,..?,,,,, (KWIKSTAT 4.1 & SURVAN V2.0) run Kaplan-Meier estimation, log-rank test Cox s proportional hazards model., Medical Biostatistics, ;,,. log-rank test,, ( p < 0.01 ),, Cox s regression model.,,.,,,., III

44 ,., 88,,,.,,.,,., run.,, enter data analyze data,., 2001 Epi K-M curve;,., SPSS, statistical analysis. K-M curve pattern ( actuarial method ), sample censoring independence, time interval uniformity, censored value, data plot pattern sample size. log- rank test, censored value. Biostatistics data management, IV

45 !,,,,,,!,,. INDEPENDENT & SIGNIFICANT VARIABLE,,! V

分级和气管是否受侵是影响预后的独立危险因素关键词状腺肿瘤 ; 癌, 乳头状 ; 预后 Clinical outcomes of 600 papillary thyroid carcinoma patients ZHANG Zong-min,

分级和气管是否受侵是影响预后的独立危险因素关键词状腺肿瘤 ; 癌, 乳头状 ; 预后 Clinical outcomes of 600 papillary thyroid carcinoma patients ZHANG Zong-min, 移性鳞癌患者的临床资料, 比较不同治疗手段的效果 50 例患者男性 44 例, 女性 6 例, 年龄 29~73 岁, 中位年龄 52.5 岁 ; 临床分期 :N2c 占 64% (32/50),N3 占 36% (18/50) 30 例患者行双颈淋巴结清扫术 50 例患者均行放疗, 剂量为 50~66