~22-5347/96/1552-0587$03.00/0 THE JOURNAL OF UROLOGY
Vol. 155,587489, February 1996
Copyright 0 1996 by AMERICAN UROLOCEAL. ASSOCIATION, INC.
Printed in U . S A
THE MANAGEMENT OF PATIENTS WITH CLINICAL STAGE I NONSEMINOMATOUS TESTICULAR TUMORS AND PERSISTENTLY ELEVATED SEROLOGIC MARKERS SCOTT B. SAXMAN, CRAIG R. NICHOLS, RICHARD S. FOSTER, JONATHAN E. MESSEMER, JOHN P. DONOHUE AND LAWRENCE H. EINHORN From the Departments of Medicine (Division of HematologylOncology) and Urology, Indiana University School of Medicine, Indianapolis, Indiana
Purpose: We reviewed our experience with patients who had nonseminomatous germ cell tumors clinically limited to the testis and persistently elevated senun human chorionic gonadotropin (HCG) or a-fetoprotein (AFP) levels after orchiectomy. Materials and Methods: All patients had clinical stage I disease with persistently elevated tumor markers that were not decreasing in accordance with the expected metabolic decay rate at retroperitoneal lymph node dissection. Results: Of 30 patients identified 3 had elevated AFP,24 had elevated HCG and 3 had elevation of both markers. Of the 6 patients with elevated AFP with or without ConCulTent HCG elevation 5 (83%)had relapse and required chemotherapy, as did 6 of 24 (25%) with HCG elevation. Conclusions: Patients with persistently elevated AFP after orchiectomy should be treated initially with chemotherapy. Although the majority of patients with elevated serum HCG were disease-free after surgery alone, a fourth of these patients still had relapse and required chemotherapy. KEY WORDS: testicular neoplasms; lymph node excision; retroperitoneal neoplasms; tumor markers, biological
The predictable pattern of spread as well as advances in radiographic imaging techniques have improved the accuracy of clinical staging in patients with testicular cancer. Patients without evidence of disease outside the testis on physical examination, and with normal abdominal and chest computerized tomography (CT) are categorized as having clinical stage I disease. The majority of these patients have normal serum levels of human chorionic gonadotropin (HCG) and a-fetoprotein (AFP).However, occasionally patients with nonseminomatous germ cell tumors clinically limited to the testis have persistently elevated serum HCG or AFP after orchiectomy. Historically these patients have undergone nerve sparing retroperitoneal lymph node dissections based on the assumption that the elevated marker is an indicator of residual disease confined to the retroperitoneal lymph nodes.' However, Davis et al recently suggested that these patients are at high risk for recurrence after retroperitoned lymph node dissection alone and should be treated initially with chemotherapy.2 We reviewed our experience with patients undergoing retroperitoneal lymph node dissection who had persistently elevated tumor markers after orchiectomy to define the risk of recurrence and optimal management. PATIENTS AND METHODS
Between January 1981 and January 1994, 573 patients with clinical stage I nonseminomatous germ cell tumors of the testis underwent retroperitoneal lymph node dissection at our university, of whom 100 patients (17%) had elevated Serum markers at operation. Serum AFP and HCG levels were determined by standard radioimmunoassay techniques. All values greater than 25 ng./ml. for AFP and greater than
2.0 mIU/ml. for HCG were considered abnormal. In all patients physical examination did not demonstrate any evidence of metastatic disease. CT of the abdomen and CT or whole lung tomography of the chest were normal. From these 100 patients, those with persistently elevated tumor markers that were not decreasing in accordance with the expected metabolic decay rate were identified. Serum tumor markers were measured from specimens drawn either immediately before or after orchiectomy and immediately before retroperitoneal lymph node dissection, and the metabolic decay rate was calculated with the formula C = 100 x 2-",where C is the percent of initial concentration and n is the number of half-lives. For purposes of this calculation the normal halflife of AFP was assumed to be 6 days, and the half-life of HCG was 36 Patients in whom both sets of markers were not evaluated and those who received adjuvant chemotherapy after retroperitoneal lymph node dissection were excluded from the study. Of the 100 patients seen during this period 20 received adjuvant chemotherapy after retroperitoneal lymph node dissection, 30 had serum markers that were decreasing per the expected decay rate, 17 did not have 2 sets of markers evaluated and old records were not available for 3. Thus, 30 patients met the aforementioned criteria and are the subject of this report. Patient age ranged from 17 to 42 years (median age 29). Median followup was 36 months (range 7 to 114). Three patients had elevated AFP only, 24 had elevated HCG only and 3 had elevation of both markers. AFP elevation ranged h m 26.5 to 1,110 ngJml. (median 98)and HCG ranged from 2.3 to 358 mIUlml. (median 10.2). Relapse was defined as an increasing serum marker after retroperitoneal lymph node dissection, or the appearance of radiographic or clinical evidence of disease.
Accepted for publication Auguat 4, 1995. Supported in part by Outstanding Investigator Grant 4682771. RESULTS Editor's Note: Thie article ie the fifth of 6 publish4 hthis issue for which catekory 1 CME c-ta eap be e-d. InAu 30 patients underwent retroperitoneal lymph node disstructions for &credits are gwen mth the qU-0section within 6 weeks after initial diagnosis and radical on pages 718 and 717. 587
ELEVATED MARKERS AFTER ORCHIECTOMY FOR STAGE I TESTIS CANCER
malignant causes of persistent HCG elevations, including hypogonadism and marijuana US^.^.^ It often is possible distinguish nonmalignant causes of HCG elevation by giving the patient an injection of testosterone and reevaluating the serum marker level in 1 to 2 weeks. Normalization of HCG &r testosterone injection is presumptive evidence of a false-positive elevation but this is not universally true. As 811 example, 1 of our patients had an HCG level of 39.1mIU/ml. before retroperitoneal lymph node dissection and a postoperative level of 52 mIU/ml. The retroperitoneal lymph nodes were pathologically negative. After an injection of testosterone the HCG was still elevated at 36.1 mIU/ml. He was followed closely without further therapy and HCG remained elevated but stable for 4 years without evidence of disease. In clinical practice, patients with clinical stage I disease and normal serum markers are either observed expectantly (surveillance)7 or undergo retroperitoneal lymph node dissection. However, a few patients have clinical stage I disease with persistently elevated serum markers (that is serologic disease only) after orchiectomy. Treatment of these patients has been controversial. Davis et a1 recently examined 11 patients with elevated markers at retroperitoneal lymph node dissection and found elevation of HCG only in 6, AFP only in 3 and both markers in 2.2All 11 patients had persistent marker elevation following retroperitoneal lymph node dissection and were treated with chemotherapy. Davis et al concluded that all patients with persistently elevated markers after orchiectomy should be treated initially with chemotherapy. It is not entirely certain, however, whether all of these patients had persistent or systemic disease after retroperitoneal lymph node dissection and were destined to suffer relapse. Some patients may have had persistent marker elevation due to other nonmalignant causes, which may account for the difference in relapse rates between their study DISCUSSION Testicdar cancer is one of the few malignancies for which and oms since we define disease persistence or relapse as an relatively specific serologic tumor markers are available.4 increasing serum marker rather than a persistently elevated serum marker after retroperitoneal lymph node dissection. and HCG are useful diaenosticallv as well as for monitoring patients receiving chemotherapy and detecting early relapse after surgery or chemotherapy. A steadily increasing CONCLUSIONS serum marker is evidence of persistent or recurrent disease even in the absence of disease detectable by radiographic We reviewed our experience with patients who had clinical methods or physical examination. However, there are non- stage I nonseminomatous testicular cancer and persistently
orchiectomy. Of the 24 patients with HCG elevation 7 had pathologically positive lymph nodes in the retroperitoneum, while 17 had negative nodes (table 1).Embryonal cell carcinoma was found in 6 of the former 7 patients, including 3 who also had teratoma in the lymph node specimen. The remaining patient with positive lymph nodes had seminoma only. The number of positive nodes ranged from 1 to 8 (median 1). In 21 of the 24 patients serum HCG normalized after retroperitoneal lymph node dissection. Of the remaining 3 patients 2 had persistent elevation of HCG postoperatively, including 1 with progression of disease, 1 who is disease-free with no subsequent therapy and 1 in whom postoperative markers were not measured. Overall 6 of the 24 patients (25%)had progressive disease 2 to 84 months after retroperitoneal lymph node dissection. Of these 6 patients 4 had pathologically positive lymph nodes at operation (1,4,4 and 8 positive nodes) and 2 with relapse had pathologically negative retroperitoneal nodes. Of the 3 patients with AFP elevation 2 had pathologically negative retroperitoneal lymph nodes and 1had seminoma in a single retroperitoneal node (table 2). All 3 patients had progressive disease within 10 months after retroperitoneal lymph node dissection (range 1 to 10). Of the 3 patients with both markers elevated 2 had pathologically negative retroperitoneal lymph nodes and 1 had embryonal carcinoma in a single retroperitoneal node. Two of these patients had relapse 2 months and 12 months after retroperitoneal lymph node dissection, respectively, while 1 was disease-free at 26 months postoperatively. Of the 6 total patients with AFP elevation the markers in 3 normalized after retroperitoneal lymph node dissection, 2 had persistently elevated AFP and 1 was lost to followup.
TABLE1. Marker level data and followuu of Datients with Persistently elevated HCG Pt. No.
Pre-Retroperitoneal Lymph Node Dissection
1 Less than 25 2.3 363 2 34.1* 3.7 3,950 3 373: 3.9 187 4 738 49.2' 4.9 5 29.2" 5.5 3.690 6 Less than 25 6.3 807 7 Less than 25 6.9 1.989 8 Less than 25 8.8 26 9 190 Less than 25 9.9 10 9.9 Less than 25 5.0 11 196 31' 10.0 12 10.2 5,700 Less than 25 13 1.750 10.4 Less than 25 14 Less than 25 635 11.0 15 Less than 2.0 Less than 25 12.1 16 149.3* 2,316 15.6 17 16.9 2,740 Less than 25 18 1.518 36.1 39' 19 284 39.1 Less than 25 20 Less than 25 1.738 40.2 21 11 30.0 Less than 25 261 81.8 22 Less than 25 23 4.700 116 61.5' 24 2,893 358 Less than 25 Serum marker elevated but decreasing according to the expeeted half-life.
Pos. Nodes (pathology)
8 (embryonal cell Ca) 0 0 0 0 0 0
0 1 (embryonal cell Ca, teratorna)
0 1 (embryonal cell Ca, teratoma) 4 (embryonal cell Ca) 0 0 0
1 (seminoma) 0 0 0
4 (embryonal cell Ca, teratorna) 1 (embryonal cell Ca) 0 0
Relapse (34) No evidence of disease (13) No evidence of disease (21) No evidence of disease (36) No evidence of disease ( 5 6 ) No evidence of disease (36) No evidence of disease (37) No evidence of disease (14) Relapse (84) No evidence of disease ( 13 1 No evidence of disease 148) Relapse (2) No evidence of disease ( 4 8 ) Relapse (3) No evidence of disease (38) No evidence of disease (7) No evidence of disease (1141 No evidence of disease (201 No evidence of disease (491 No evidence o f disease (24) Relapse (9) No evidence of disease I37) No evidence of disease ( 5 7 ) Relapse (1)
ELEVATED MARKERS AFTER ORCHIECTOMY FOR STAGE I TESTIS CANCER
TABLE2. Marker level data and followup in patients with persistently elevated AFP or without elevation of HCG preoperatively HCG (mIU/ml.) AFP ~mTu/ml.) Pos. Nodes (pathology) Statua (ma.)
Pt. No. 25 26 27 28 29 30
Less than 2.0 Less than 2.0 Less than 2.0 7.8 3.7 27.6
1 semin no ma)
1,110 35 69 90.7 26.5 127.5
0 0 1 (embryonal cell Ce) 0 0
Relapse (1) Relapse (9) Relapse (10) Relapse (2) Relapse ( 12) No evidence of disease (26)
elevated markers a t retroperitoneal lymph node dissection. minimize morbidity while maximizing the chances for cure, This is clearly a select group of patients with clinical stage I the treatment plan should be individualized. An option is disease, since it does not include those with normal markers, close surveillance with repeat measurement of blood markmarkers decreasing according to the expected half-life or ers, monthly chest radiographs and frequent abdominal CT. treatment with adjuvant chemotherapy after retroperitoneal Patients with increasing markers or radiographic evidence of lymph node dissection. However, of the 6 patients with ele- progressive disease should be treated promptly with chemovated AFP with or without concurrent elevation in HCG 5 therapy, while those with stable or decreasing markers can had relapse and required chemotherapy, which suggests that continue to be observed closely. Another option for patients they should be treated primarily with chemotherapy rather with more marked elevations in HCG would be chemotherthan retroperitoneal lymph node dissection, with a cure rate apy as primary treatment, with surgery reserved for those that should approach 100%. with recurrent teratoma in the retroperitoneum. Still anOf the 24 patients with elevated serum HCG alone 75% other option would be to repeat the HCG measurement aRer were long-term disease-free survivors after retroperitoneal administration of testosterone. If HCG subsequently normallymph node dissection only. It is likely that this figure actu- izes the patient could be offered retroperitoneal lymph node ally underestimates the number of patients who would have dissection as is our current practice for those with clinical relapse after retroperitoneal lymph node dissection alone, stage I disease and normal serum markers. since it excludes those who received adjuvant chemotherapy. Because all patients treated with adjuvant chemotherapy REFERENCES have pathological lymph node involvement it is reasonable to 1. Donohue, J. P., Thornhill,J. A., Foster, R. S., Rowland, R. G. and assume that more of them would have relapse than in this Bihrle, R.: Retroperitoneal lymphadeneetomyfor clinical stage group of 24, of whom 17 had pathologically negative findings. A testis cancer (1965 to 1989):modifications of technique and It is also difficult to explain why the serum HCG normalized impact on ejaculation. J. Urol., 149: 237, 1993. after retroperitoneal lymph node dissection in 15 of these 2. Davis, B. E., Herr, H. W., Fair, W. R. and Bosl, G. J.: The patients despite the fact that they did not have pathologically management of patients with nonseminomatous germ cell tumors of the testis with serologic disease only aRer orchiecinvolved retroperitoneal lymph nodes. It is possible that this tomy. J. Urol., 152 111, 1994. finding was due to sampling errors and that some of these 3. Einhorn, L. H.,Richie, J. P. and Shipley, W.U.: Cancer of the patients had small areas of lymph node involvement that testis. In: Cancer: Principles and Practice of Oncology, 4th ed. were undetected. A more probable explanation is that occaEdited by V. T. DeVita, S. Hellman and S. A. Rmeuberg. Philsionally patients have false-positive elevations in HCG beadelphia: J. B. Lippincott Co., p. 1130, 1993. fore retroperitoneal lymph node dissection and are already 4. Bod, G. J., Geller, N. L., Cimncione, C., Nisselbaum, J., Vugrin, disease-free after orchiectomy. It is likely that the risk of D., Whitmore, W. F.,Jr. and Golbey, R. B.: Serum tumor having persistent disease (that is the true positive rate) markers in patienta with metastatic germ cell tumors of the testis. A 10-year experience. Amer. J. Med., 75: 29, 1983. correlates with the magnitude of HCG elevation. However, 5. Phillips, M. I., Weiner, R., MacLaren, N. and Kappy, M.: Test of because of the relatively small sample size no statistical false-positivetesticular cancer marker by suppressing luteinstatements can be made regarding the degree of HCG elevaizing hormone with testosterone. Lancet, 2: 928, 1982. tion and the probability of the test being false-positive. 6. Laehrer, P. J., Eisenhut, C. and Sample, M.: How to cope with Overall, these data suggest that retroperitoneal lymph antibodies to HCG in laboratory testing. J. CLin. Ligand Assay node dissection has therapeutic benefits in a minor% Soc., 10: 58, 1987. these patients. Of the 24 patients who underwent ~ t r o p e n - 7. Read, G., Stenning, S. P., Cullen, M. H., Parkinson, M. C., toneal lymph node dissection 17 had negative lymph nodes Horwich. A., Kaye, S. B. and Cook, P. A: Medical Research Council prospective study of surveillance for stage I testicular and presumably were not helped by the surgery, and 4 ofthe teratoma. Medical Research Council Testicular Tumors Work7 Patients with positive nodes had relapse and required !heing Party. J. Clin. Oncol., 10: 1762, 1992. motherapy. Since the goal of therapy in these patients 1s to