The Second Medical Research Council Study of Prognostic Factors in Nonseminomatous Germ Cell Tumors By G.M. Mead, S.P. Stenning, M.C. Parkinson, A. Horwich, S.D. Fossa, P.M. Wilkinson, S.B. Kaye, E.S. Newlands, and P.A. Cook for the Medical Research Council Testicular Tumour Working Party Purpose: To assess prognostic factors in a large population of patients with metastatic nonseminomatous germ cell tumors (NSGCT) arising in gonadal or extragonadal sites. Patients and Methods: Data from 795 patients treated with chemotherapy between 1982 and 1986 in 13 centers were analyzed. Particular emphasis was placed on exact tumor measurements (eg, size of nodal masses, number of lung metastases), and the diagnostic pathology was also reviewed. Cox regression analysis was performed on these data. The patients were treated with a variety of cisplatin-containing chemotherapy regimens, 86% of which included etoposide. Results: With median follow-up of 45 months, overall 3-year survival is 85%. The independently adverse features proved to be (1) the presence of liver, bone, or brain metastases; (2) raised marker levels (alpha-fetoprotein
IN
[AFP] level > 1,000 kU/L or beta subunit of human chorionic gonadotropin [HCG] > 1,000 IU/L); (3) the presence of a mediastinal mass greater than 5 cm in diameter; (4) the presence of 20 or more lung metastases; (5) increasing age; and (6) absence of undifferentiated teratoma (embryonal carcinoma) or fibrous tissue from the primary tumor. Conclusions: The first four factors were used to define a simple prognostic classification. A good-prognosis group having none of these features comprised 67% of our patient population and had a 3-year survival of 93%. The remaining 33% of patients having at least one of these features had a 3-year survival rate of 68%. These patient groups are currently the subjects of international randomized clinical trials. J Clin Oncol 10:85-94. © 1992 by American Society of Clinical Oncology. PATIENTS AND METHODS
1985 THE MEDICAL Research Council (MRC)
published a multivariate analysis of prognostic factors in 458 patients with metastatic nonseminomatous germ cell tumors (NSGCT) treated in six centers between 1976 and 1982.' This analysis demonstrated the
independent importance of tumor volume (assessed by
the Royal Marsden Hospital staging system2) and degree of elevation of the tumor markers alpha-fetoprotein (AFP) and human chorionic gonadotropin (HCG) in determining prognosis. Both of these factors were related to delay in diagnosis, which was identified as an important and potentially reversible determinant of long-term survival. In addition, the analysis showed age and calendar year of treatment to be independent
prognostic factors. The decrease in the proportion of patients with more advanced disease in successive years
did not fully explain the apparent improvement in survival among patients treated in the later years of the
study. This improvement was attributed to increased experience of the participants in the management of this uncommon disease. Although copies of the original
Eligibility Participating clinicians were asked to provide data on all patients of the following three types treated by chemotherapy in their centers between June 1982 (the closing date of the previous MRC study) and June 1986: (1) patients with metastatic NSGCT from a testicular primary; (2) patients with primary extragonadal NSGCT; and (3) patients with metastatic testicular seminoma and elevated AFP levels (> 25 kU/L). ClinicalData Collection Clinical data were collected retrospectively on a form designed for the study. Prechemotherapy information collected included serum AFP and HCG levels in the week before treatment and lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBD) levels where available. The largest transverse diameters of masses in the retroperitoneum, mediastinum, and neck were recorded and the diameter of the largest lung mass (data taken preferentially from computed tomographic [CT] scan) was noted together with a count of the total number of lung metastases. The presence of liver, bone, or brain metastases was recorded. Stage according to the Royal Marsden Hospital system and volume as defined in the previous study' were computed from this raw data.
histopathology reports were obtained, central pathology review was not performed in this study. The current study, which included central histopatho-
logic review, was conducted to assess prognosis in a large patient group in the more recent treatment era, and
specifically, to devise a simple prognostic classification on which to base different therapeutic approaches.
From the Medical Research Council, London, England. Submitted March 4, 1991; acceptedAugust 5, 1991. Address reprint requests to G.M. Mead, DM, MRCP, Royal South Hants Hospital,Graham Rd, Southampton, S09 4PE, England. © 1992 byAmerican Society of Clinical Oncology. 0732-183X/92/1001-0017$3.00/0
Journalof Clinical Oncology, Vol 10, No 1 (January), 1992: pp 85-94
Downloaded from ascopubs.org by UNIVERSITY LIVERPOOL on April 20, 2019 from 154.059.124.102 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
85
86
MEAD ET AL
Details of chemotherapy given were recorded, together with previous use of radiotherapy. Response to treatment, with residual disease defined as a mass at least 2 cm in diameter, was assessed according to the following six categories: (1) Complete remission (CR): all patients in this category had returned to normal serum AFP and HCG levels. Where residual masses were present, these were completely resected and found to contain either differentiated teratoma (mature somatic elements), fibrosis, or necrosis only. (2) Partial remission (PR)-cancer: patients in this category had either persistent elevation of AFP or HCG level at the end of therapy associated with a greater than 50% reduction in the size of tumor masses, or complete resection of residual masses found to contain viable malignancy (malignant teratoma, undifferentiated/ embryonal carcinoma, trophoblastic tumor, and yolk sac tumor). (3) PR-residual mass: patients in this category had returned to normal levels of AFP or HCG. Residual masses present were either watched, incompletely resected, or treated with radiotherapy. (4) No response/progressive disease: failure of serum markers to normalize and associated with no response or growth of tumor masses. (5) Toxic death. (6) Unassessable. Date of relapse/ progression (after initial chemotherapy plus surgery) in addition to date of death or last visit were also recorded. The completed forms were returned to the MRC Cancer Trials Office in Cambridge, England, where data management was performed using COMPACT. 3
RESULTS Treatment Only 3% of patients had received previous radiotherapy. A range of chemotherapy regimens was used,' the most common of which were bleomycin, etoposide, and cisplatin (BEP; 38%); cisplatin, vincristine, methotrexate, and bleomycin (POMB)/dactinomycin, cyclophosphamide, and etoposide (ACE) (16%); cisplatin, vinblastine, and bleomycin (PVB; 14%); cisplatin, vinblastine, and etopside (PVE; 11%); and BEP/PVB (7%). Only
one patient failed to receive primary cisplatin-based therapy, and 86% received combinations containing both cisplatin and etoposide (compared with 54% in the previous MRC analysis). The percentage of patients receiving both cisplatin and etoposide increased from 62% in 1982 to 93% in 1986. Problems of patient selection make comparison of specific regimens difficult to interpret. However, there was no evidence that patients receiving PVB' and other non-etoposidecontaining regimens had lower survival rates than those patients treated with regimens containing both cisplatin and etoposide.
HistopathologicReview
Overall Results
A complete set of sections cut from the blocks originally selected from the primary tumor of each patient was requested. Each neoplasm was assessed in terms of pathologic staging and tumor classification in the manner described previously4 and by one histopathologist (M.C.P.) without knowledge of the clinical course.
Seven hundred ninety-five patients from 13 centers were entered onto the study (Table 1). The median follow-up time was 45 months. Follow-up of 2 years was available in 96% of patients, and 76% of patients were observed for 3 years or more. The overall 3-year survival was 85% (95% confidence interval [CII, 82% to 88%), and progression-free survival was 82% (95% CI, 79% to 85%). One hundred thirty patients (16%) are known to have died: 106 died due to uncontrolled malignancy, 17 due to chemotherapyrelated toxicity, three after surgery for residual masses,
StatisticalMethods The dates of the study were set to allow a minimal follow-up time of 2 years in a sufficient number of patients to make survival a viable end point. Response to treatment was recorded but was not assessable in many cases because residual masses were unresected. Thus, survival was the principal end point used in this analysis, measured from the date of commencement of chemotherapy. Because of interlaboratory variation in the measurement of LDH and HBD, the normal ranges of these variables were obtained from each center and analysis was performed according to whether or not the level for each patient was more than twice the upper limit of normal within that center. Other continuous variables were categorized according to recognized groupings (eg, the Royal Marsden staging system for diameter of lymph nodes) and assessed using the log-rank test 5 for comparing survival curves for different levels of each of the features. For simplicity, categories with similar survival rates were combined (so that each became a binary variable). All variables achieving a log-rank P value less than 0.1 were included in multivariate analyses to identify independently important factors. Cox's proportional hazards regression model6 was used with the statistical package BMDP7 and a forward stepwise variable selection procedure. All X2 statistics quoted are on 1 df unless otherwise stated.
Table 1. Contributing Centers Center
No. of Patients
Bath Birmingham Charing Cross Cambridge Clatterbridge Glasgow Leeds The London Hospital Manchester Mount Vernon NRH, Oslo, Norway Royal Marsden Southampton
8 37 77 40 31 81 48 49 94 14 102 181 33
Abbreviation: NRH, Norwegian Radium Hospital.
Downloaded from ascopubs.org by UNIVERSITY LIVERPOOL on April 20, 2019 from 154.059.124.102 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
87
PROGNOSTIC FACTORS IN NSGCT
three due to cardiac failure, and one due to lung cancer. In the final evaluation, 614 patients (77%) were alive and disease-free, 45 (6%) were alive with stable or diminishing residual masses, and six were alive with disease. According to the response categories, 414 (52%) patients achieved a CR on chemotherapy alone, and a further 123 (16%) required surgery to achieve CR. Ninety-five patients (12%) had a PR-cancer, and 131 (16%) had a PR-residual mass. Thirteen patients (2%) showed no response. Of the remaining 19 patients, 18 died before the response assessment (14 from toxic deaths), and one did not attend follow-up evaluation. The outcome in individual patients in each response category is summarized in Table 2. Of the patients who achieved a CR, 95% are alive and free of active disease; this percentage fell to 70% for those with unresected residual masses at the end of chemotherapy and 60% for those who achieved a PR (with markers raised at the end of chemotherapy or resected masses containing viable tumor). Seven hundred twenty-three (91%) patients had stage IM or metastatic disease from a testicular primary tumor, 53 (7%) had extragonadal NSGCT (33 retroperitoneal primaries, 11 mediastinal primaries, and nine of unknown primary site), and 19 had seminomas of the testis with raised AFP and metastatic disease.
As listed in Table 3, 72% of patients had paraaortic node metastases, 45% had lung metastases visible on chest x-ray or chest CT, 9% had supraclavicular spread, and 8% had mediastinal disease. In addition, 9% of patients had liver metastases and 4% had bone or CNS involvement. Ten percent of patients had raised markers as the only sign of advanced disease. Overall, 65% were classified as having small-volume disease comprising stages IM, IIA and B, IIIA and B, and IVLL 2,A,B, compared with only 50% in the series of patients entered into the last MRC study of prognostic factors. Sixteen percent had large-volume disease, and 18% had very large-volume disease. Markers. Table 5 gives 3-year survival rates for various levels of AFP, HCG, LDH, and HBD. Both AFP and HCG levels were significantly related to length of survival, with HCG showing the strongest association. The combined variable marker (defined as low if AFP level was < 1,000 kU/L and HCG level was < 10,000 IU/L, and high if at least one of the markers was raised above this level) identified a group of 589 patients with an AFP level less than 1,000 kU/L and an HCG level less than 10,000 IU/L with a 3-year survival rate of 91%. Individually, both LDH and HBD levels were significantly associated with length of survival. However, readings were available on only 34% of all patients, and it was, therefore, not possible to include these variables in the subsequent Cox regression analysis. Other clinical variables. Table 6 lists other clinical variables investigated. Patients with extragonadal primary tumors appeared to have a poorer prognosis than patients with testicular primary tumors. Of the 33 patients with retroperitoneal primaries, 12 have died; five of 11 patients with mediastinal primaries and three
PrognosticFactors-UnivariateAnalyses Staging variables. The 3-year survival rates are listed for different levels of each of the individual staging variables (Table 3) and for volume (as defined in the previous MRC study) and number of sites of metastases 9 (Table 4). All variables were related to survival at a high level of statistical significance (P < .01; log-rank test).
Table 2. Final Outcome by Response to Treatment Alive, Residual Mass Stable
Alive and in CR
CR After chemotherapy only After chemotherapy + surgery PR-cancer Viable cancer resected Elevated markers only PR-residual mass Residual mass unresected No response Early death/unassessable Total
Alive, Active Disease
Dead
No.
%
No.
%
No.
%
387 118
93 96
2 2
2 2
3
1
-
22 28
58 60
1 1
3 2
1
2
58
41
28
5
2 -
1
1
39 -
-
614
45
-
6
No.
%
Total
22 3
5 2
414 123
15 17
39 36
38 47
42 13 18
30 100 95
141 13 19
130
Downloaded from ascopubs.org by UNIVERSITY LIVERPOOL on April 20, 2019 from 154.059.124.102 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
795
88
MEAD ET AL Table 4. Summary Measures of Extent of Disease
Table 3. 3-Year Survival Rates for Staging Variables
Paraaortic nodes (cm) None
IN
[AFP] level > 1,000 kU/L or beta subunit of human chorionic gonadotropin [HCG] > 1,000 IU/L); (3) the presence of a mediastinal mass greater than 5 cm in diameter; (4) the presence of 20 or more lung metastases; (5) increasing age; and (6) absence of undifferentiated teratoma (embryonal carcinoma) or fibrous tissue from the primary tumor. Conclusions: The first four factors were used to define a simple prognostic classification. A good-prognosis group having none of these features comprised 67% of our patient population and had a 3-year survival of 93%. The remaining 33% of patients having at least one of these features had a 3-year survival rate of 68%. These patient groups are currently the subjects of international randomized clinical trials. J Clin Oncol 10:85-94. © 1992 by American Society of Clinical Oncology. PATIENTS AND METHODS
1985 THE MEDICAL Research Council (MRC)
published a multivariate analysis of prognostic factors in 458 patients with metastatic nonseminomatous germ cell tumors (NSGCT) treated in six centers between 1976 and 1982.' This analysis demonstrated the
independent importance of tumor volume (assessed by
the Royal Marsden Hospital staging system2) and degree of elevation of the tumor markers alpha-fetoprotein (AFP) and human chorionic gonadotropin (HCG) in determining prognosis. Both of these factors were related to delay in diagnosis, which was identified as an important and potentially reversible determinant of long-term survival. In addition, the analysis showed age and calendar year of treatment to be independent
prognostic factors. The decrease in the proportion of patients with more advanced disease in successive years
did not fully explain the apparent improvement in survival among patients treated in the later years of the
study. This improvement was attributed to increased experience of the participants in the management of this uncommon disease. Although copies of the original
Eligibility Participating clinicians were asked to provide data on all patients of the following three types treated by chemotherapy in their centers between June 1982 (the closing date of the previous MRC study) and June 1986: (1) patients with metastatic NSGCT from a testicular primary; (2) patients with primary extragonadal NSGCT; and (3) patients with metastatic testicular seminoma and elevated AFP levels (> 25 kU/L). ClinicalData Collection Clinical data were collected retrospectively on a form designed for the study. Prechemotherapy information collected included serum AFP and HCG levels in the week before treatment and lactate dehydrogenase (LDH) and hydroxybutyrate dehydrogenase (HBD) levels where available. The largest transverse diameters of masses in the retroperitoneum, mediastinum, and neck were recorded and the diameter of the largest lung mass (data taken preferentially from computed tomographic [CT] scan) was noted together with a count of the total number of lung metastases. The presence of liver, bone, or brain metastases was recorded. Stage according to the Royal Marsden Hospital system and volume as defined in the previous study' were computed from this raw data.
histopathology reports were obtained, central pathology review was not performed in this study. The current study, which included central histopatho-
logic review, was conducted to assess prognosis in a large patient group in the more recent treatment era, and
specifically, to devise a simple prognostic classification on which to base different therapeutic approaches.
From the Medical Research Council, London, England. Submitted March 4, 1991; acceptedAugust 5, 1991. Address reprint requests to G.M. Mead, DM, MRCP, Royal South Hants Hospital,Graham Rd, Southampton, S09 4PE, England. © 1992 byAmerican Society of Clinical Oncology. 0732-183X/92/1001-0017$3.00/0
Journalof Clinical Oncology, Vol 10, No 1 (January), 1992: pp 85-94
Downloaded from ascopubs.org by UNIVERSITY LIVERPOOL on April 20, 2019 from 154.059.124.102 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
85
86
MEAD ET AL
Details of chemotherapy given were recorded, together with previous use of radiotherapy. Response to treatment, with residual disease defined as a mass at least 2 cm in diameter, was assessed according to the following six categories: (1) Complete remission (CR): all patients in this category had returned to normal serum AFP and HCG levels. Where residual masses were present, these were completely resected and found to contain either differentiated teratoma (mature somatic elements), fibrosis, or necrosis only. (2) Partial remission (PR)-cancer: patients in this category had either persistent elevation of AFP or HCG level at the end of therapy associated with a greater than 50% reduction in the size of tumor masses, or complete resection of residual masses found to contain viable malignancy (malignant teratoma, undifferentiated/ embryonal carcinoma, trophoblastic tumor, and yolk sac tumor). (3) PR-residual mass: patients in this category had returned to normal levels of AFP or HCG. Residual masses present were either watched, incompletely resected, or treated with radiotherapy. (4) No response/progressive disease: failure of serum markers to normalize and associated with no response or growth of tumor masses. (5) Toxic death. (6) Unassessable. Date of relapse/ progression (after initial chemotherapy plus surgery) in addition to date of death or last visit were also recorded. The completed forms were returned to the MRC Cancer Trials Office in Cambridge, England, where data management was performed using COMPACT. 3
RESULTS Treatment Only 3% of patients had received previous radiotherapy. A range of chemotherapy regimens was used,' the most common of which were bleomycin, etoposide, and cisplatin (BEP; 38%); cisplatin, vincristine, methotrexate, and bleomycin (POMB)/dactinomycin, cyclophosphamide, and etoposide (ACE) (16%); cisplatin, vinblastine, and bleomycin (PVB; 14%); cisplatin, vinblastine, and etopside (PVE; 11%); and BEP/PVB (7%). Only
one patient failed to receive primary cisplatin-based therapy, and 86% received combinations containing both cisplatin and etoposide (compared with 54% in the previous MRC analysis). The percentage of patients receiving both cisplatin and etoposide increased from 62% in 1982 to 93% in 1986. Problems of patient selection make comparison of specific regimens difficult to interpret. However, there was no evidence that patients receiving PVB' and other non-etoposidecontaining regimens had lower survival rates than those patients treated with regimens containing both cisplatin and etoposide.
HistopathologicReview
Overall Results
A complete set of sections cut from the blocks originally selected from the primary tumor of each patient was requested. Each neoplasm was assessed in terms of pathologic staging and tumor classification in the manner described previously4 and by one histopathologist (M.C.P.) without knowledge of the clinical course.
Seven hundred ninety-five patients from 13 centers were entered onto the study (Table 1). The median follow-up time was 45 months. Follow-up of 2 years was available in 96% of patients, and 76% of patients were observed for 3 years or more. The overall 3-year survival was 85% (95% confidence interval [CII, 82% to 88%), and progression-free survival was 82% (95% CI, 79% to 85%). One hundred thirty patients (16%) are known to have died: 106 died due to uncontrolled malignancy, 17 due to chemotherapyrelated toxicity, three after surgery for residual masses,
StatisticalMethods The dates of the study were set to allow a minimal follow-up time of 2 years in a sufficient number of patients to make survival a viable end point. Response to treatment was recorded but was not assessable in many cases because residual masses were unresected. Thus, survival was the principal end point used in this analysis, measured from the date of commencement of chemotherapy. Because of interlaboratory variation in the measurement of LDH and HBD, the normal ranges of these variables were obtained from each center and analysis was performed according to whether or not the level for each patient was more than twice the upper limit of normal within that center. Other continuous variables were categorized according to recognized groupings (eg, the Royal Marsden staging system for diameter of lymph nodes) and assessed using the log-rank test 5 for comparing survival curves for different levels of each of the features. For simplicity, categories with similar survival rates were combined (so that each became a binary variable). All variables achieving a log-rank P value less than 0.1 were included in multivariate analyses to identify independently important factors. Cox's proportional hazards regression model6 was used with the statistical package BMDP7 and a forward stepwise variable selection procedure. All X2 statistics quoted are on 1 df unless otherwise stated.
Table 1. Contributing Centers Center
No. of Patients
Bath Birmingham Charing Cross Cambridge Clatterbridge Glasgow Leeds The London Hospital Manchester Mount Vernon NRH, Oslo, Norway Royal Marsden Southampton
8 37 77 40 31 81 48 49 94 14 102 181 33
Abbreviation: NRH, Norwegian Radium Hospital.
Downloaded from ascopubs.org by UNIVERSITY LIVERPOOL on April 20, 2019 from 154.059.124.102 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
87
PROGNOSTIC FACTORS IN NSGCT
three due to cardiac failure, and one due to lung cancer. In the final evaluation, 614 patients (77%) were alive and disease-free, 45 (6%) were alive with stable or diminishing residual masses, and six were alive with disease. According to the response categories, 414 (52%) patients achieved a CR on chemotherapy alone, and a further 123 (16%) required surgery to achieve CR. Ninety-five patients (12%) had a PR-cancer, and 131 (16%) had a PR-residual mass. Thirteen patients (2%) showed no response. Of the remaining 19 patients, 18 died before the response assessment (14 from toxic deaths), and one did not attend follow-up evaluation. The outcome in individual patients in each response category is summarized in Table 2. Of the patients who achieved a CR, 95% are alive and free of active disease; this percentage fell to 70% for those with unresected residual masses at the end of chemotherapy and 60% for those who achieved a PR (with markers raised at the end of chemotherapy or resected masses containing viable tumor). Seven hundred twenty-three (91%) patients had stage IM or metastatic disease from a testicular primary tumor, 53 (7%) had extragonadal NSGCT (33 retroperitoneal primaries, 11 mediastinal primaries, and nine of unknown primary site), and 19 had seminomas of the testis with raised AFP and metastatic disease.
As listed in Table 3, 72% of patients had paraaortic node metastases, 45% had lung metastases visible on chest x-ray or chest CT, 9% had supraclavicular spread, and 8% had mediastinal disease. In addition, 9% of patients had liver metastases and 4% had bone or CNS involvement. Ten percent of patients had raised markers as the only sign of advanced disease. Overall, 65% were classified as having small-volume disease comprising stages IM, IIA and B, IIIA and B, and IVLL 2,A,B, compared with only 50% in the series of patients entered into the last MRC study of prognostic factors. Sixteen percent had large-volume disease, and 18% had very large-volume disease. Markers. Table 5 gives 3-year survival rates for various levels of AFP, HCG, LDH, and HBD. Both AFP and HCG levels were significantly related to length of survival, with HCG showing the strongest association. The combined variable marker (defined as low if AFP level was < 1,000 kU/L and HCG level was < 10,000 IU/L, and high if at least one of the markers was raised above this level) identified a group of 589 patients with an AFP level less than 1,000 kU/L and an HCG level less than 10,000 IU/L with a 3-year survival rate of 91%. Individually, both LDH and HBD levels were significantly associated with length of survival. However, readings were available on only 34% of all patients, and it was, therefore, not possible to include these variables in the subsequent Cox regression analysis. Other clinical variables. Table 6 lists other clinical variables investigated. Patients with extragonadal primary tumors appeared to have a poorer prognosis than patients with testicular primary tumors. Of the 33 patients with retroperitoneal primaries, 12 have died; five of 11 patients with mediastinal primaries and three
PrognosticFactors-UnivariateAnalyses Staging variables. The 3-year survival rates are listed for different levels of each of the individual staging variables (Table 3) and for volume (as defined in the previous MRC study) and number of sites of metastases 9 (Table 4). All variables were related to survival at a high level of statistical significance (P < .01; log-rank test).
Table 2. Final Outcome by Response to Treatment Alive, Residual Mass Stable
Alive and in CR
CR After chemotherapy only After chemotherapy + surgery PR-cancer Viable cancer resected Elevated markers only PR-residual mass Residual mass unresected No response Early death/unassessable Total
Alive, Active Disease
Dead
No.
%
No.
%
No.
%
387 118
93 96
2 2
2 2
3
1
-
22 28
58 60
1 1
3 2
1
2
58
41
28
5
2 -
1
1
39 -
-
614
45
-
6
No.
%
Total
22 3
5 2
414 123
15 17
39 36
38 47
42 13 18
30 100 95
141 13 19
130
Downloaded from ascopubs.org by UNIVERSITY LIVERPOOL on April 20, 2019 from 154.059.124.102 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
795
88
MEAD ET AL Table 4. Summary Measures of Extent of Disease
Table 3. 3-Year Survival Rates for Staging Variables
Paraaortic nodes (cm) None
