Medical Research Council Prospective Study of Surveillance for Stage I Testicular Teratoma By G. Read, S.P. Stenning, M.H. Cullen, M.C. Parkinson, A. Horwich, S.B. Kaye, and P.A. Cook for the Medical Research Council Testicular Tumors Working Party Purpose: A prospective study of surveillance after orchidectomy alone in patients with stage I nonseminomatous germ cell testicular tumor (NSGCT) was performed to determine the relapse-free rate and to identify the histologic criteria that predict for relapse. Patients and Methods: Three hundred ninety-six patients from 16 United Kingdom and one Norwegian centers were entered onto the study between January 1, 1984 and October 1, 1987 of whom 373 were eligible for analysis. In a previous retrospective study, we defined a prognostic index based on histologic criteria that identified a group of patients with a high risk of relapse. This index was based on the presence of venous and lymphatic invasion, undifferentiated cells, and the absence of yolk sac elements in the primary tumor.

SEVERAL

STUDIES HAVE established that surveillance after orchidectomy alone for stage I nonseminomatous germ cell testicular tumor (NSGCT) is a satisfactory alternative method of management to retroperitoneal node dissection (RPLND) with chemotherapy for relapsed patients. 1-7 Radiotherapy (RT) was previously used in the United Kingdom as a primary treatment after orchidectomy. Although it has been shown to be effective in controlling low-volume retroperitoneal nodal disease7,8 it may compromise the success of chemotherapy in patients who subsequently relapse; therefore, it is no longer recommended. In 1987, the Medical Research Council (MRC) reported a retrospective study of 259 patients from 10 collaborating centers in the United Kingdom. 9 In this study, the relapse-free rate at 2 years was 74%, with an estimated 4-year survival rate of 98%. A subgroup of approximately 20% of patients was identified who had a high risk of relapse as determined by histologic criteria. To establish the relapse rate in the longer term (after 18 months of follow-up), the pattern of relapse, and to verify in an independent data set the histologic criteria previously identified, a prospective multicenter study of surveillance of stage I NSGCT treated by orchidectomy alone was begun in 1984. This report is based on those patients who were registered before October 1987 when a phase

From the MedicalResearch Council,London, United Kingdom. Submitted February18, 1992; acceptedJune 22, 1992. Address reprint requests to Graham Read, MD, Christie Hospital, Wilmslow Rd, Withington, ManchesterM20 9BX, United Kingdom. © 1992 by American Society of Clinical Oncology. 0732-183X/92/1011-0015$3.00/0

1762

Results: The 2-year actuarial relapse-free rate after orchidectomy was 75% (95% confidence interval, 71% to 79%), and the rate at 5 years was 73%. Five patients died of tumor or treatment-related complications, which resulted in a 5-year survival of 98%. The relapse-free rate in patients with three or four risk factors was 54%. Conclusions: This study confirms the safety of surveillance as a method of management and identifies a group of patients with a high risk of relapse. A prospective phase II study has been initiated to determine whether two courses of platinum-based adjuvant chemotherapy will prevent relapse in these high-risk patients. J Clin Oncol 10:1762-1768. © 1992 by American Society of Clinical Oncology.

II study of adjuvant chemotherapy for high-risk stage I NSGCT was introduced. PATIENTS AND METHODS Three hundred ninety-six patients from 16 United Kingdom and one Norwegian centers were entered onto the prospective study (Table 1). All patients had histologically verified NSGCT, or seminoma with increased alphafetoprotein (AFP), which was treated initially by orchidectomy alone between January 1, 1984 and October 1, 1987. Patients were staged by clinical examination, chest x-ray, computed tomographic (CT) scanning of the thorax and abdomen, and assay of AFP and human chorionic gonadotropin (HCG) levels. Lymphography and ultrasound investigations were optional. The criterion for entry was that patients had stage I disease determined by the Royal Marsden staging system, and patients satisfying this criteria were registered by means of a telephone call to the MRC Cancer Trials Office, Cambridge. Patients with tumor at the cut end of the spermatic cord and those who were unavailable for follow-up were not registered. Patients with equivocal CT scans or raised markers were registered, but deemed ineligible if repeat scans were found positive or if the marker levels never returned to normal. Nineteen patients were ruled ineligible on these grounds, three were excluded because they had received previous RT for seminoma of the contralateral testis, and one patient was excluded for having a pure seminoma without raised markers, which left 373 eligible patients for analysis. Follow-up attendances after staging investigations were to be at monthly intervals for the first year, every 2 months for the second, and every 3 months in the third. Assessment consisted of clinical examination, chest x-ray, and AFP and HCG assay at each visit. Marker investigations were performed more frequently if there was clinical need. CT scanning was performed at each center according to one of two schedules. In schedule A, scanning of the thorax and abdomen was performed at alternate visits for the first year and every 6 months in the second year. In schedule B, scanning of the thorax and abdomen was performed at 2, 4, 8, and 12 months. Scans were not performed routinely after 2 years, but were

Journalof Clinical Oncology, Vol 10, No 11 (November), 1992: pp 1762-1768

Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.

SURVEILLANCE FOR STAGE I NSGCT

1763

Table 1. Number of Patients at Each Center No. of Patients

Center

Addenbrookes Hospital, Cambridge Christie Hospital, Manchester Churchill Hospital, Oxford Clatterbridge Hospital, Merseyside Cookridge Hospital, Leeds Derby Royal Infirmary, Derby Leicester Royal Infirmary, Leicester London* Mount Vernon Hospital, Middlesex Norwegian Radium Hospital, Oslo, Norway Nottingham General Hospital, Nottingham Queen Elizabeth Hospital, Birmingham Royal Marsden Hospital, Sutton Royal South Hants Hospital, South Hants Western Infirmary, Glasgow, Scotland Western Park Hospital, Sheffield York District Hospital, York Total

27 87 11 2 26 8 6 23 12 10 6 17 83 20 38 14 6 396

*Incorporates the London Hospital, the Institute of Urology, and Shaftesbury Hospital.

performed on other occasions according to clinical indication. Six centers opted for option A and 11 for option B. The initial staging investigations were repeated on identification of relapse. The treatment of relapse was dependent on the stage and followed the usual practice for each participating center. Histologic sections from orchidectomy specimens were reviewed by the reference pathologist (M.C.P.). The classification and staging systems used were those described in our previous retrospec9 tive study. All data forms were sent to the MRC Cancer Trials Office in Cambridge where they were verified and computerized using the COMPACT data management package (Timberlake Clark Ltd, London, UK). 1o

StatisticalMethods No specific sample size was set when the study was initiated, but an initial accrual of at least 100 patients was anticipated to assess the prognostic index that was to be derived from the retrospective study.

9

Relapse-free interval was taken from the date of orchidectomy. Relapse-free rates were calculated using the Kaplan-Meier method and compared using the log-rank test." Multivariate analyses used 12 Cox's proportional hazards regression model, using forward and backward stepwise variable selection procedures. From histopathologic data collected in the retrospective MRC study of surveillance 9 in stage I teratoma, a prognostic index was derived that identified a group of patients at high risk of relapse. The index was based on the number of the following features displayed: invasion of testicular veins, invasion of testicular lymphatics, absence of yolk sac elements, and presence of undifferentiated cells. The high-risk group displayed any three or all four of these features. This classification was tested prospectively on this independent data set.

RESULTS The median follow-up for those patients who were relapse-free when last seen was 5 years. A minimum of 2

years follow-up was available on all but five patients (currently lost to follow-up and censored at 4, 6, 19, 22, and 23 months after orchidectomy). One hundred patients (27%) have relapsed, 80% in the first 12 months after orchidectomy. Table 2 shows the timing of relapse, and the overall relapse-free curve is shown in Fig 1. The annual risk of relapse seemed to decrease sharply after the first 12 months following orchidectomy, and continued to decrease in the second and subsequent years. The 2-year actuarial relapse-free rate was 75% (95% confidence interval [CI], 71% to 79%), and decreased to 73% 5 years after orchidectomy (95% CI, 68% to 78%). The latest relapse to have occurred in this group of patients was 42 months after orchidectomy. Additionally, four patients have developed tumors of the contralateral testis (6, 9, 24, and 31 months from orchidectomy). In each case after a second orchidectomy, the patients returned to a policy of surveillance. Royal Marsden stage13 at relapse is detailed in Table 3. By the criteria established in a previous MRC study,14 nine patients had large-volume disease and two had very large-volume disease. Thirteen percent of patients had raised markers that were the only sign of relapse, but of the 87 patients with metastatic disease, 39 (45%) did not have raised markers when their relapse was diagnosed. The paraaortic nodes were the most common site of relapse (61% of cases). One quarter of patients relapsed with lung metastases. Relapse in the mediastinal (7%) and supraclavicular (3%) nodes was less common. One patient had liver metastases and one had brain metastases. Postrelapsetreatment. The four patients who developed contralateral tumors were treated by second orchidectomy. The histology was seminoma in each case. Of the 100 relapsed patients, one received surgery only for an inguinal recurrence, and the remaining 99 patients received chemotherapy according to the usual practice of the referring center. Residual masses after chemotherapy were removed surgically whenever possiTable 2. Rate of Relapse Time (months)

0-6 7-12 13-18 19-24 25-30 31-36 37-48 49-60 61-72

No. of Relapses

No. at Risk During Period

Annual Relapse Rate (%)

54 26 6 6 4 2 1 1 0

373 318 290 284 275 270 258 232 202

29 16 4 4 3 1 3 weeks

No.

2-Year Relapse-Free Rate

22 236 94 12

95 77 64 83

9.5 on 3 df P = .02

215 158

71 80

4.1 on 1 df P = .04

194 179

72 79

2.9 on 1 df P = .09

Log-Rank X2

Abbreviations: TTP, testicular tumor panel; UICC, International Union Against Cancer; df, degree of freedom; MTI; malignant teratoma intermediate; MTU, malignant teratoma undifferentiated; TD, teratoma differentiated.

2 187

-

2

-

75

28

61

93

83

5.08 on 1 df

5.8 on 2 df P = .05

6.6 on 3 df P = .09

52

88

299

72

4.23 on 1 df P = .04

120 246

72 76

0.47 on 1 df P = .49

41

93

325

73

P = .03

96 270

65 78

5.16 on 1 df P = .02

106

67

260

78

4.2 on 1 df P = .04

177 188

69 80

4.84 on 1 df

4.3 on 1 df

P = .04

invasion was less in comparison to the first study, their inclusion reduced the relapse-free rate in the high-risk group from 65%, if defined by presence of venous invasion alone, to 54% if all four factors are used. Risk factors and relapse characteristicsin retrospective andprospective study patients. The ability of the individual risk factors to predict relapse at a particular site (for example, abdominal node relapse in patients with lymphatic invasion; lung, liver, bone, or brain metastases in patients with venous invasion) was examined in both the retrospective and prospective study patients. Because similar results were observed in the two groups, they were combined to give greater statistical power to detect such trends. Thus the population available comprised a total of 162 relapsed patients for whom pathology review was available: 62 from the retrospective study and 100 from the prospective study. Each of the individual risk factors, together with high (at least three features) or low (at most two features) risk group, were examined in relation to relapse in the abdominal, mediastianal, or supraclavicular nodes; lung, liver, bone, or brain; and relapse with or without raised markers, to Royal Marsden stage at relapse, and to the time of relapse. There was a nonsignificant trend for patients with venous invasion to relapse more frequently in the lung and for the high-risk patients to have fewer late (> 9 months from orchidectomy) relapses. The risk of nodal relapse was not increased in patients with lymphatic invasion, and there were no overall differences in stage among high- and low-risk groups. A significant negative association between abdominal relapse and venous invasion was noted; approximately three quarters of patients without venous invasion relapsed in the abdominal nodes compared with half of those with venous invasion

Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.

1767

SURVEILLANCE FOR STAGE I NSGCT

(X2 = 8.1 on 1 df; P = .005). This should be interpreted cautiously in view of the multiple comparisons being performed. Thus there is no real support for a further tailoring of the surveillance policy on the basis of the histopathologic assessment of the primary tumor. DISCUSSION

This series of 373 patients is the largest reported series of prospectively identified stage I NSGCT patients who were treated by orchidectomy alone. The 2-year actuarial relapse-free rate is 75% (95% CI, 71% to 79%), and this only decreases to 73% 5 years after orchidectomy. The results are similar to those of our retrospective study that, recently updated with a median follow-up of 54 months, showed a 2-year relapse-free rate of 74% (95% CI, 69% to 79%) and a 5-year rate of 71% (95% CI, 65% to 77%). The pattern of relapse supports a policy of intensive surveillance for the 2-year period after orchidectomy, but continued follow-up is necessary to 5 years to detect late relapses. Eighty-nine (89%) patients had small-volume disease on relapse by the MRC criteria. This confirms the effectiveness of the surveillance policy. Two patients currently have active disease, but the remainder were rendered disease-free or have residual masses under surveillance after chemotherapy. Four patients have relapsed and died and another has died from bleomycin lung toxicity. Thus only five patients have died of tumor or treatment-related complications. Three had smallvolume disease at the time of relapse and one largevolume, and failure of treatment cannot, therefore, be attributed to the surveillance policy. Two patients have active disease at the time of this report and currently are undergoing chemotherapy. The survival is again similar to our reported retrospective study and equals the best published series for RPLND with salvage chemotherapy.16-20 The recently updated retrospective study continues to show a 98% survival rate. The present study again confirms the effectiveness of the histologic prognostic index reported in our retrospective study as a means of selecting a group of patients with a high probability of relapse who are suitable for inclusion in studies of adjuvant chemotherapy. Multivariate analysis showed venous invasion to be the most important individual variable giving a high-risk group with a relapse-free rate of 65%. This accords with other published studies. 21 However some series do not distinguish between venous and lymphatic invasion 5 or feel that such a pathologic distinction is not possible. 22-25 In our series, only nine of 61 (15%) of patients with

invasion of the testicular lymphatics did not also show venous invasion. The presence of undifferentiated cells (embryonal carcinoma) has been reported as an adverse prognostic factor in some series5, 24,26 but not others. 6,21,25 The absence of yolk sac elements may be artifactual because of the more accurate identification of patients who are truly stage I. That is, that tumors that contain yolk sac elements have a greater propensity to produce AFP, and a more reliable detection of metastases is possible in patients with persistently raised AFP. The SWENOTECA group 27 (Swedish-Norwegian project on nonseminanatous testicular germ-cell cancer) found that raised preorchidectomy AFP was an independent factor that predicted retroperitoneal metastases in clinical stage I patients, and that when this variable was omitted, absence of yolk sac tumor became predictive of retroperitoneal metastases. In a series reported by Hoskin et al, 13 the presence of an increased serum concentration of AFP preorchidectomy was associated with a lower relapse rate. The addition of lymphatic invasion, the presence of undifferentiated cells, and the absence of yolk sac elements improved the prognostic index over venous invasion alone, which allowed definition of a high-risk group with a relapse-free rate of 54%. Thus we do not propose any change in our pathologic index. An analysis of risk factors and relapse characteristics showed a trend for patients with venous invasion to relapse in the lungs and a statistically significant negative association with abdominal relapse. However, the association was not felt to be sufficiently strong to recommend any further tailoring of the surveillance policy in the low-risk patients. The incidence of contralateral testicular tumor in this study was 1% (four of 373), which is in accordance with reported series of 1% to 2%.28,29 The tumor was seminoma in each case, and the patients have remained well after orchidectomy alone. In conclusion, this study confirms the effectiveness of surveillance as a policy for stage I NSGCT of the testis with a 5-year survival rate of 98%. It also confirms the findings of our previous report that a group of patients with a high probability of relapse that comprise one quarter of the total can be identified on histologic criteria. A prospective phase II study of adjuvant chemotherapy has now been initiated to determine whether two courses of chemotherapy given immediately after orchidectomy in patients with high-risk pathologic features when the volume of disease is minimal will be 30 sufficient to prevent relapse.

Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.

1768

READ ET AL APPENDIX

Medical Research Council Testicular Tumors Working Party members are as follows: J.P. Blandy, R.B. Buchanan, M.A. Cornbleet, M.H. Cullen, S.D. Fossa, E.D. Gilby, J.A. Green, S.J. Harland,W.F. Hendry, W. G. Jones, M. King, F. Madden, G.M. Mead, A. Horwich (chairman), R.T.D. Oliver, E.S. Newlands, M.C. Parkinson (reference pathologist), R.M. Phillips, G. Read, M.R.G. Robinson, G.J.S. Rustin, S.P. Stenning (statistician), P. Wilkinson, M V. Williams. Other study participantsinclude: C. Alcock, R. Ashford, A. Benghiat, N.M. Bleehen, C.H. Brown, A. Champion, H.J. Close, P. Corbett, S. Dische, K. Durrant, D. Fermont, A. Jones, A. Laing, M. Peckham, B. Richards, M. Sokal, T. Wheeler, J.M.A. Whitehouse, and H. Yosef. P.A. Cook and E. Lallemand (data managers). (Italics indicate current study participants.)

REFERENCES 1. Peckham MJ, Barrett A, Husband JE, et al: Orchidectomy alone for stage 1 testicular non-seminoma. Lancet 2:678-680, 1983 2. Read G, Johnson RJ, Wilkinson PM, et al: Prospective study of follow up alone in stage 1 teratoma of the testis. Br Med J 287:1503-1505, 1983 3. Pizzocaro G, Zanoni F, Milani A, et al: Orchidectomy alone in clinical stage 1 non-seminomatous testis cancer: A critical appraisal. J Clin Oncol 4:35-40, 1986 4. Raghaven D, Coils B, Levi J, et al: Surveillance for stage I non-seminomatous germ cell tumours of the testis: The optimal protocol has not yet been defined. Br J Urol 61:522-526, 1988 (abstr 532). 5. Sogani PC, Fair WR: Surveillance alone in the treatment of clinical stage I non-seminomatous germ cell tumor of the testis. Semin Urol 6:53-56, 1988 6. Thompson PI, Nixon J, Harvey VJ: Disease relapse in patients with stage I nonseminomatous germ cell tumors of the testis on active surveillance. J Clin Oncol 6:1597-1603, 1988 7. Rorth M, von der Masse H, Nielsen ES, et al: Orchidectomy alone versus orchidectomy plus radiotherapy in stage I nonseminomatous testicular cancer. A randomised study by the Danish Testicular Carcinoma Study Group. Int J Androl 10:255-262, 1987 8. Peckham MJ, Barrett A, McElwain TJ, et al: Non-seminoma germ-cell tumours of the testis: Results of treatment and an analysis of prognostic factors. Br J Urol 53:162-172, 1981 9. Freedman LS, Parkinson MC, Jones W, et al: Histopathology in the prediction of relapse of patients with stage 1 testicular teratoma treated by orchidectomy alone: A Medical Research Council Collaborative study. Lancet 2:294-298, 1987 10. COMPACT steering committee: Improving the quality of data in clinical trials in cancer: The COMPACT computer package. Br J Cancer 63:412-415, 1990 11. Peto R, Pike MC, Armitage P, et al: Design and analysis of randomised clinical trials requiring prolonged observation of each patient. II. Analysis. Br J Cancer 35:1-39, 1977 12. Tibshirani R: A plain man's guide to the proportional hazards model. Clin Invest Med 5:63-68, 1982 13. Hoskin P, Dilly S, Easton D, et al: Prognostic factors in stage I non-seminomatous germ cell testicular tumors managed by orchidectomy and surveillance: Implications. J Clin Oncol 4:10311036, 1986 14. Medical Research Council Working Party on Testicular Tumours: Prognostic factors in advanced non-seminomatous germcell testicular tumours: Results of a multicentre study. Lancet 1:8-11, 1985 15. Read G, Johnson RJ, Wilkinson PM: The role of radiotherapy after chemotherapy in the management of persistent paraaortic disease in non-seminomatous germ cell tumours. Br J Cancer 53:623-628, 1986

16. Donohoe JP: Retroperitoneal lymphadenopathy: The anterior approach including bilateral supra-renal hilar dissection. Urol Clin N Am 44:509-523, 1977 17. Staubitz WJ, Early KS, Magoss IV, et al: The surgical treatment of non-seminomatous germinal testis tumors. Cancer 32:1206-1211, 1973 18. Williams SD, Stablein DM, Einhorn LH, et al: Immediate adjuvant chemotherapy versus observation with treatment at relapse in pathological stage II testicular cancer. N Engl J Med 317:1433-1438, 1987 19. Fossa SD, Klepp O, Ous S, et al: Unilateral retroperitoneal lymph node dissection for testicular cancer. J Urol 127:685-688, 1982 20. Richie JP: Modified retroperitoneal lymphadenectomy for patients with clinical stage I testicular cancer. Semin Urol 6:216222, 1988 21. Fung CY, Kalish LA, Brodsky GL, et al: Stage I nonseminomatous germ cell testicular tumor: Prediction of metastatic potential by primary histopathology. J Clin Oncol 6:1467-1473, 1988 22. Pont J, Hotl W, Kosak D, et al: Risk-adapted treatment choice in stage I nonseminomatous testicular germ cell cancer by regarding vascular invasion in the primary tumour: A prospective trial. J Clin Oncol 8:16-20, 1990 23. Holtl W, Kosak D, Pont J, et al: Testicular cancer: Prognostic implications of vascular invasion. J Urol 137:683-685, 1987 24. Fujime M, Chang H, Lin CW, et al: Correlation of vascular invasion and metastasis in germ cell tumours of the testis: A preliminary report. J Urol 131:1237-1241, 1984 25. Rodriguez PN, Hafez GR, Messing EM: Non-seminomatous germ cell tumour of the testicle: Does extensive staging of the primary predict the likelihood of metastatic disease? J Urol 136:604-608, 1986 26. Dunphy CH, Ayala AG, Swanson DA, et al: Clinical stage I nonseminomatous and mixed germ cell tumors of the testis. A clinicopathological study of 93 patients on a surveillance protocol after orchidectomy alone. Cancer 62:1202-1206, 1988 27. Klepp O, Olsson AM, Henrikson H, et al: Prognostic factors in clinical stage I nonseminomatous germ cell tumors of the testis: Multivariate analysis of a prospective multicenter study. J Clin Oncol 8:509-518, 1990 28. Pugh RCB: Pathology of the testis. Oxford, England Blackwell Scientific Publications, 1976 29. Sokal M, Peckham MJ, Hendry WF: Bilateral germ cell tumours of the testis. Br J Urol 52:158-162, 1980 30. MRC Cancer Trials Office: Protocol TEO5-Treatment of 'high risk' stage I testicular teratoma, 1987

Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.

Medical Research Council prospective study of surveillance for stage I testicular teratoma. Medical Research Council Testicular Tumors Working Party.

A prospective study of surveillance after orchidectomy alone in patients with stage I nonseminomatous germ cell testicular tumor (NSGCT) was performed...
638KB Sizes 0 Downloads 0 Views

Recommend Documents