|
|||||
|
|
||||||
© 2000 American Society for Clinical Oncology Biologic Factors Determine Prognosis in Infants With Stage IV Neuroblastoma: A Prospective Childrens Cancer Group StudyFrom the Department of Pediatrics, University of Illinois at Chicago College of Medicine, Chicago, IL; Department of Pediatrics, University of California School of Medicine, San Francisco; Departments of Pediatrics, Pathology, and Preventive Medicine, University of Southern California School of Medicine and Childrens Hospital; University of California Los Angeles School of Medicine, Los Angeles; The Childrens Cancer Group, Arcadia, CA; Department of Pediatrics, University of Pennsylvania School of Medicine, Philadelphia, PA; Department of Pediatrics, Vanderbilt University, Nashville, TN; and Denver Pediatric Surgeons, Denver, CO. Address reprint requests to Mary Lou Schmidt, MD, Childrens Cancer Group, PO Box 60012, Arcadia, CA, 91066-6012; email mls3{at}uic.edu
PURPOSE: A prospective Childrens Cancer Group study, CCG-3881, has been completed to determine if a more accurate prediction of prognosis by biologic features can identify subgroups of infants with stage IV neuroblastoma (NBL) who require differing intensities of treatment. PATIENTS AND METHODS: One hundred thirty-four infants were registered from June 1989 to August 1995, with a median follow-up of 47.1 months (range, 0 to 88 months). The biologic factors examined were tumor MYCN copy number, Shimada histopathologic classification, serum ferritin, and bone marrow immunocytology (sensitivity, one tumor cell per 105 bone marrow cells). Patients treated on CCG-3881 (n = 116) received four-drug chemotherapy for 9 months (cisplatin, cyclophosphamide, doxorubicin, and etoposide), with surgery and local radiation to residual disease. After January 1991, all subsequent infants with tumor MYCN amplification (n = 18) were transferred after one cycle of therapy to the high-risk CCG-3891 protocol (open January 1991 to April 1996) for more intensive treatment. RESULTS: The 3-year event-free survival (EFS) and overall survival (mean ± SD) for the 134 infants were 63% ± 5% and 71% ± 5%, respectively. Patients whose tumors were without MYCN amplification had a 93% ± 4% 3-year EFS, whereas those with amplified MYCN had a 10% ± 7% 3-year EFS (P < .0001). Each of the other biologic features studied had prognostic significance in univariate analysis but not after stratifying by MYCN copy number.
CONCLUSION: Infants less than 1 year of age at diagnosis with stage IV NBL have a much improved outcome compared with children
INFANTS LESS THAN 1 year of age at diagnosis with stage IV neuroblastoma (NBL) have a much better prognosis than older children with stage IV NBL, with a 50% to 75% event-free survival (EFS) compared with a less than 30% EFS for older children.1-8 Nevertheless, some patients experience rapid disease progression despite intensive therapy. Biologic features, including tumor MYCN gene amplification, Shimada histopathology, serum ferritin, bone marrow immunocytology (BMI), and DNA content have been shown to have prognostic value in NBL.9-20 The purpose of this study was to identify subgroups of infants with stage IV NBL with significantly different probabilities of survival.
Patient Population All infants with Evans stage IV NBL were registered onto Childrens Cancer Group (CCG) trial CCG-3881 (open June 1989 to August 1995).21 After 1991, the high-risk NBL protocol, CCG-3891, opened and an amendment was written to CCG-3881 to transfer all infants with tumor MYCN amplification and disseminated disease to CCG-3891 because of these patients proven high risk for failure shown in the first 13 such patients treated on CCG-3881. Infants were transferred to CCG-3891 after the first course of chemotherapy when the MYCN determination became available. All patients entered onto either study had an informed consent signed by a parent or guardian and study approval by the appropriate local institutional human research review board.
Biologic Features
Treatment
Statistical Analysis
Patients A total of 134 infants with Evans stage IV NBL were registered onto CCG-3881 and 3891. Table 1 lists biologic data, including serum ferritin for 86% of patients, MYCN for 76% of patients, BMI for 61%, and Shimada histopathology for 68% of patients. All four biologic features were measured in 28% (n = 37) of the patients. At least three of these four features were measured in 71% (n = 95) of the patients; at least two of the four in 93% (n = 124); and at least one of the four in 100% (n = 134). Unfavorable characteristics were seen in approximately one third of the patients; 30% of patients had MYCN amplification, 31% had unfavorable pathology, 33% had elevated ferritin, and 44% had positive BMI.
Local Tumor Control and EFS Best eventual surgical resection (complete resection [CR], microscopic residual [MR], partial resection [PR], and biopsy only [BX]) was analyzed (Fig 1). More extensive surgical resection, CR or MR, was not associated with an improved EFS compared with lesser resection, PR or BX. After stratification for nonamplified MYCN tumors, the 3-year EFS (mean ± SD) for those with gross resection (CR + MR) was 91% ± 5% compared with 94% ± 6% for those with residual tumor (PR + BX) (P = .72). Patients with MYCN-amplified, grossly resected tumors had a 3-year EFS of 14% ± 9% (CR + MR) compared with 10% ± 10% for those with residual tumor (PR + BX) (P = .18) (Table 2).
EFS and Biologic Variables The 3-year EFS for all infants with stage IV NBL treated on CCG-3881 and CCG-3891 was 63% ± 5%, and the overall survival (OS) was 71% ± 5% (median follow-up, 47.1 months; range, 0 to 88 months) (Fig 2A and 2B). When comparing patients with nonamplified MYCN tumors with patients whose tumors showed amplification of MYCN, the difference in 3-year EFS (93% ± 4% v 10% ± 7%, respectively) was highly significant (P < .0001) (Fig 3A). There was no difference in the 3-year EFS for patients with amplified MYCN tumors treated on CCG-3881 (EFS = 8% ± 5%) versus CCG-3891 (EFS = 11% ± 7%). Because 24% of patients in this analysis are missing MYCN data, it is possible that missing data is concealing the importance of other biologic or clinical features in the children that we have studied here. In fact, overall, patients whose MYCN copy number data were unknown (3-year EFS = 51% ± 10%) had outcome marginally worse than patients who had known MYCN copy number (3-year EFS = 67% ± 6%) (P = .07). This difference, however, was restricted to the initial phase of the studies. Before January 1991, before the high failure rate of infants whose tumors showed MYCN amplification was established, these infants received moderate intensity therapy according to CCG-3881. Forty-one percent of these early patients did not have MYCN copy number determined, and the EFS for these patients was significantly worse than for those patients with known MYCN data treated at the same time. By January 1991, it was recognized that infants with disseminated NBL whose tumors showed MYCN amplification had a very poor prognosis. CCG-3881 was amended and all such infants were transferred to the high-risk CCG-3891 protocol. After January 1991, there were only 18% of patients missing MYCN copy number, and the EFS for patients missing MYCN data was identical to that of patients with known copy number. In this later time period (after January 1991), MYCN remained highly prognostic, with a 3-year EFS of 11% ± 6% for patients with MYCN-amplified tumors and a 3-year EFS of 95% ± 4% for nonamplified tumors (P < .0001).
In addition, Shimada histopathology, serum ferritin, and bone marrow involvement as measured by BMI were each significant in univariate analysis of EFS (Fig 3B, 3C, and 3D). Age 0 to 3 months or greater than 3 months was not significant (Table 2). In the multivariate Cox analysis of each biologic factor relative to MYCN copy number, no biologic variable was identified that enhanced the prognostic value of MYCN copy number (Table 2).
Toxicity
Progression and Death
Recent CCG studies have identified a number of potentially important prognostic factors and biologic characteristics for children with NBL.9,15,21 These include MYCN copy number, the Shimada histopathologic classification, serum ferritin levels, and BMI at diagnosis. Previously, these factors have been examined primarily in the advanced stages of NBL (stages III and IV) in children greater than 1 year of age, but the prognostic significance of all these biologic variables for infant stage IV disease has not been adequately evaluated. Before the study of biologic characteristics of NBL tumors, many investigators recognized that the progression-free survival rate of 50% to 80% for infants less than 1 year of age with stage IV NBL was more favorable than for older children.1-7 In CCG-371, 43 infants with stage IV disease received combined modality treatment with cyclophosphamide, vincristine, dacarbazine ± doxorubicin, teniposide, and cisplatin, which resulted in a 5-year disease-free survival of 49%. Twenty-four infants treated at the Dana-Farber Cancer Institute/Boston Childrens Hospital (Boston, MA) from 1970 to 1988 with multiagent chemotherapy had a 5-year actuarial survival rate of 75%.3 Treatment with 4 months of cyclophosphamide and doxorubicin at St Jude Childrens Research Hospital (Memphis, TN) of 44 infants with disseminated NBL, including 12 stage IV-S patients, resulted in an 80% survival.4 In a recent report from the Pediatric Oncology Group, 62 stage D patients received five courses of cyclophosphamide and doxorubicin, with or without cisplatin, and had a 60% 5-year actuarial survival rate.6 In none of these previous reports were biologic factors studied to define prognostic subgroups of infants who would be at high risk for relapse. Lampert et al24 noted that infants with disseminated NBL whose tumors showed hyperdiploidy, nonamplification of MYCN, normal chromosome 1p, and high serum and tumor levels of cellular matrix protein CD44 had an excellent prognosis. All patients with MYCN amplification progressed. It was not shown whether the additional analysis of ploidy or CD44 was prognostic independent of MYCN status. The aim of CCG-3881 was to refine the prognostic classification of more favorable NBL by combining the clinical variables of stage and age with the biologic variables, including MYCN gene status, Shimada histopathology, serum ferritin, and BMI. The results of the treatment of infants with stage IV NBL on CCG-3881 and CCG-3891 show that MYCN copy number is the most important predictor of prognosis in this group. Patients with MYCN-amplified tumors had a 3-year EFS of 10% compared with a 93% 3-year EFS for patients with nonamplified MYCN tumors. Despite the fact that 24% of the patients did not have MYCN copy number measured, the overall EFS for the patients with unknown MYCN copy number was not significantly different from that of patients who had known MYCN copy number. Because MYCN remained so highly prognostic in the later time period (after the protocol change in January 1991) when little MYCN data was missing, we suspect that other variables must play, at most, a subordinate role to MYCN amplification for prediction of EFS in patients with stage IV disease. Of the 41 patients who died, only one patient had documented nonamplified MYCN copy number; whereas 28 patients had amplified MYCN tumors, and 12 had tumors with unknown MYCN copy number. After stratification by tumor MYCN copy number, no other biologic feature examined (histopathology, ferritin, or BMI) or clinical feature (surgical resection, age 0 to 3 months v > 3 months, or chemotherapy regimen) altered prognosis. This is consistent with two previous reports that assessed ploidy, CD44 values and/or clinical features, and MYCN copy number. In these, MYCN copy number retained significance independent of all variables or was only analyzed in a minimum number of cases.20,24 Genetic characteristics have recently been reported as having prognostic significance. The Pediatric Oncology Group has demonstrated the prognostic significance of the DNA content of NBL for infants.18-20 In their most recent study of 172 infants < 1 year of age with unresectable NBL, 56 had stage D disease.20 Thirty-three tumors showed hyperdiploidy, and 23 showed diploid DNA content. For the entire group of infants with unresectable NBL, hyperdiploidy was associated with a 94% 3-year OS, and diploid DNA content was associated with a 55% OS. In this study, MYCN amplification was strongly associated with diploid DNA content (odds ratio of 4.6, P < .01), and after stratification, MYCN copy number retained prognostic significance for both hyperdiploid and diploid tumors. (This report did not examine the statistical significance of diploid DNA content for nonamplified MYCN tumors in stage IV infants.) Current intergroup studies in the United States (which are investigating MYCN copy number, Shimada histopathology, and DNA content for all infants < 12 months of age) expect a yearly accrual of 50 infants with stage IV NBL, two out of three of whom are expected to have tumors with single-copy MYCN. Our data, in which only 7% of patients with nonamplified MYCN tumors relapsed (two to three per year among expected accrual), suggest that future studies may take as long as 10 years to decide whether or not other prognostic variables (such as DNA content) are predictive in this subset of infants (assuming a relative risk of 3). Our data indicate a poor outcome (10% 3-year EFS, 12% OS) for infants with MYCN amplification. On the other hand, the Bowman et al20 study found that, among 16 infants with MYCN-amplified tumors, an estimated 37.5% of such patients were still alive after 3 years. Although this figure applies to all stages of infants, in our own experience MYCN amplification among infants is extremely rare in stages other than stage IV. Therefore, a pooled estimate (from the Bowman and the current report) of survival for amplified stage IV infants is 21%, which evidently is within sampling variation of the estimate from either study. Thus, there is no strong evidence that the survival of MYCN-amplified stage IV patients was different for the Bowman et al20 study compared with the current one. Again, our data suggest that it will require lengthy observation to determine whether any other prognostic variable (such as DNA content) is predictive of favorable outcome among infants with MYCN-amplified stage IV NBL. In conclusion, the current study clearly shows that new approaches are required for infants less than 12 months of age with stage IV MYCN-amplified NBL in an effort to improve their current dismal prognosis. Possible approaches might include myeloablative consolidation therapy without total-body irradiation, immunotherapy, retinoids, targeted radiotherapy, or novel agents such as antiangiogenic compounds. On the other hand, chemotherapy and radiation may perhaps be safely decreased for infants with nonamplified MYCN stage IV tumors. This approach is being tested in the current intergroup study, which decreases total length of therapy for these patients relative to that given in previous CCG protocols.
Participating Principal Investigators: Childrens Cancer Group
Supported by grants from the Division of Cancer Treatment, National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Bethesda, MD.
1. Kretschmar CS, Frantz CN, Rosen EM, et al: Improved prognosis for infants with stage IV neuroblastoma. J Clin Oncol 2:799-803, 1984[Abstract] 2. Nickerson HJ, Nesbit ME, Grosfeld JL, et al: Comparison of stage IV and IV-S neuroblastoma in the first year of life. Pediatr Oncol 13:261-268, 1985 3. Paul SR, Tarbell NJ, Korf B, et al: Stage IV neuroblastoma in infants: Long-term survival. Cancer 67:1493-1497, 1991[Medline] 4. Bowman LC, Hancock ML, Santana VM, et al: Impact of intensified therapy on clinical outcome in infants and children with neuroblastoma: The St. Jude Childrens Research Hospital Experience, 1962 to 1988. J Clin Oncol 9:1599-1608, 1991[Abstract] 5. De Bernardi B, Pianca C, Boni L, et al: Disseminated neuroblastoma (stage IV and IV-S) in the first year of life: Outcome related to age and stage. Cancer 70:1625-1633, 1992[Medline] 6. Strother D, Shuster JJ, McWilliams N, et al: Results of Pediatric Oncology Group Protocol 8104 for infants with stages D and DS neuroblastoma. J Pediatr Hematol Oncol 17:254-259, 1995[Medline] 7. De Cou JM, Bowman LC, Rao BN, et al: Infants with metastatic neuroblastoma have improved survival with resection of the primary tumor. J Pediatr Surg 30:937-941, 1995[Medline] 8. Shorter NA, Davidoff AM, Evans AE, et al: The role of surgery in the management of stage IV neuroblastoma: A single institution study. Med Pediatr Oncol 24:287-291, 1995[Medline]
9.
Hann HWL, Evans AE, Siegel SE, et al: Prognostic importance of serum ferritin in patients with stages III and IV neuroblastoma: The Childrens Cancer Study Group Experience. Cancer Res 45:2843-2848, 1985
10.
Brodeur GM, Seeger RC, Schwab M, et al: Amplification of N-myc in untreated human neuroblastomas correlates with advanced disease stage. Science 224:1121-1124, 1984 11. Seeger RC, Brodeur GM, Sather H, et al: Association of multiple copies of the N-myc oncogene with rapid progression of neuroblastomas. N Engl J Med 313:1111-1116, 1985[Abstract] 12. Seeger RC, Wada R, Brodeur GM, et al: Expression of N-myc by neuroblastomas with one or multiple copies of the oncogene. Clin Biol Res 271:41-49, 1988 13. Crabbe DC, Peters J, Seeger RC, et al: Rapid detection of MYCN gene amplification in neuroblastomas using the polymerase chain reaction. Diagn Mol Pathol 1:229-234, 1992[Medline] 14. Shimada H, Chatten J, Newton WA, et al: Histopathologic prognostic factors in neuroblastic tumors: Definition of subtypes of ganglioneuroblastoma and an age-linked classification of neuroblastomas. J Natl Cancer Inst 73:405-416, 1984 15. Chatten J, Shimada H, Sather HN, et al: Prognostic value of histopathology in advanced neuroblastoma: A report from the Childrens Cancer Study Group. Hum Pathol 19:1187-1198, 1988[Medline]
16.
Shimada H, Stram DO, Chatten J, et al: Identification of subsets of neuroblastomas by combined histopathologic and N-myc analysis. J Natl Cancer Inst 87:1470-1476, 1995 17. Moss TJ, Reynolds CP, Sather HN, et al: Prognostic value of immunocytologic detection of bone marrow metastases in neuroblastoma. N Engl J Med 324:219-226, 1991[Abstract] 18. Look AT, Hayes A, Nitschke R, et al: Cellular DNA content as a predictor of response to chemotherapy in infants with unresectable neuroblastoma. N Engl J Med 311:231-235, 1984[Abstract] 19. Look AT, Hayes FA, Schuster JJ, et al: Clinical relevance of tumor cell ploidy and N-myc gene amplification in childhood neuroblastoma: A Pediatric Oncology Group Study. J Clin Oncol 9:581-591, 1991[Abstract]
20.
Bowman LC, Castleberry RP, Cantor A, et al: Genetic staging of unresectable metastatic neuroblastoma in infants: A Pediatric Oncology Group Study. J Natl Cancer Inst 89:373-380, 1997
21.
Matthay KK, Perez C, Seeger RC, et al: Successful treatment of stage III neuroblastoma based on prospective biologic staging: A Childrens Cancer Group Study. J Clin Oncol 16:1256-1264, 1998 22. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958 23. Cox DR: Regression models and life tables. J R Stat Soc B 34:187-220, 1972 24. Lampert F, Christiansen H, Berner F, et al: Disseminated neuroblastomas under 1 year of age: Cell biology and prognosis. J Neuroncol 31:181-184, 1997[Medline] Submitted October 14, 1998; accepted November 19, 1999.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||
|
Copyright © 2000 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|