Neuroblastoma: Biology and Therapy

Neuroblastoma is a clinically heterogeneous disease which can have a benign, localized behavior or a rapidly progressive, widely disseminated, lethal character. In recent years, knowledge of tumor biology—genetic changes and other biological markers—has allowed for reliable prediction of patient outcomes. These markers, when combined with clinical staging and histologic findings, determine the relative risk for a given patient and allow therapy to be tailored to that particular level of risk.

Dr. Matthay presents a clear and concise summary of the known biological markers for neuroblastoma and describes how these factors translate into a risk stratification/therapeutic schema. Loss of heterozygosity on the distal short arm of chromosome 1, amplification of the MYCN oncogene, near diploid chromosomal ploidy, low expression of the TRK-A nerve growth factor receptor, and telomerase activity each correlate closely with poor outcome. MYCN amplification and tumor cell ploidy are the most widely used markers because they are readily determined and are independent predictors of survival. When these two factors are combined with age, stage, and histopathologic grade, patients can be stratified for prognosis and therapy by risk of relapse.

Controversies in Risk Stratification

While it is not difficult to assign patients to low-risk or very-high-risk groups, defining the area of “intermediate” risk arouses controversy. The appropriateness of characterizing infants with unresectable or metastatic neuroblastoma as intermediate risk was questioned in Pediatric Oncology Group (POG) study 8743.[1] In this nonrandomized prospective study—an extension of the investigation of Look et al[2]—outcome was predicted to be correlated with tumor ploidy. Patients with hyperdiploid tumors received treatment with cyclophosphamide(Cytoxan, Neosar)/doxorubicin and were switched to cisplatin (Platinol)/teniposide (Vumon) if they responded poorly. Those with diploid tumors were immediately switched to cisplatin/teniposide and maintained on this regimen for up to eight cycles if they had an initial response. In spite of this moderately intense therapy, the overall survival rate for patients with diploid, MYCN nonamplified tumors was less than 70%.

If one defines intermediate-risk patients as those who are likely to have good survival (ie, greater than 90%) after a standard eight-cycle regimen, infants with diploid tumors do not strictly fall into this group. This result conflicts with that of previous Children’s Cancer Group (CCG) studies, as summarized in this article, which suggest that infants with single copies of MYCN have a good prognosis on standard therapy.

The next intergroup (POG/CCG) study for intermediate risk neuroblastoma will treat patients with MYCN-nonamplified diploid tumors with eight cycles of chemotherapy administered in an outpatient setting. Because the results of POG 8743 raised concerns, it will be important to examine the outcome of this study to see whether this particular subgroup is truly intermediate risk or should be moved to a high-risk regimen.[3] Some clinicians may feel that the question has already been settled and will treat these patients with a high-risk regimen off-study.

Biology and Surgery

The role of surgery in these various risk groups continues to evolve and remains controversial in some cases. While resection of the primary tumor is considered appropriate for low-risk patients (International Neuroblastoma Staging System [INSS] stage 1, infants with stage 2, 3, and 4S hyperdiploid tumors), the timing and extent of resection in high-risk patients (older children with stage 3 or 4 disease) are more problematic.

For localized neuroblastoma, negative biological markers, such as diploid DNA index and MYCN amplification, may not impact on survival,[4] and such patients may be successfully treated with complete surgical resection limited to the primary tumor only. Infants with higher stage (2, 3, 4S), hyperdiploid, MYCN-nonamplified tumors can be treated with surgery and observation alone.These tumors have a strong tendency to undergo progressive differentiation into ganglioneuromas and are unlikely to recur as biologically unfavorable metastatic tumors.

In high-risk patients with advanced disease and negative biological markers, any advantage of aggressive, up-front surgery must be balanced against the potential for complications, which could delay or alter the administration of dose-intensive chemotherapy. Although gross total excision of the primary tumor correlates with improved survival in stage 3 and 4 patients in some,[5,6] although not all[7] series, the timing of such surgery does not affect survival. Because the rate of surgical complications is higher when extensive tumors are removed at diagnosis,[8] the standard approach for advanced stage tumors has become initial biopsy (using an open or minimally invasive approach such that adequate tissue is obtained for biological studies) followed by clinical/pathologic/biological risk group assignment, chemotherapy, and second-look surgery. The goal of second-look surgery should be to remove the tumor completely without damaging adjacent structures, such as the kidney and intestines.

How tumor biology affects resectability remains an open question. MYCN amplification did not correlate either positively or negatively with resectability in a large CCG study of high-risk patients.[9] However, Nakagawa[10] and colleagues reported increased resectability of MYCN-nonamplified tumors. For MYCN-amplified tumors, complete resection led to a significant survival advantage that was not noted in the nonamplified group. What the effect of tumor biology is on resectability and how gross total resection affects survival may be definitively answered in current protocols of patients undergoing such treatment regimens.

The current therapy for neuroblastoma has emerged as a model of the integration of tumor biology and therapeutics. Although this synthesis allows for significant reductions in cytotoxic chemotherapy for patients with low-risk disease, some controversy remains regarding stratification of certain intermediate-risk patients and the extent of surgery in advanced disease. Fifty percent of all neuroblastoma patients present with high-risk disease for which no truly effective therapy yet exists. Hopefully, further improvements in our understanding of the biology of this tumor will facilitate the development of new, effective therapies.

References:

1. Bowman LC, Castleberry RP, Canter A, et al: Genetic staging of unresectable or metastatic neuroblastoma in infants: A Pediatric Oncology Group study. J Natl Cancer Inst 89:373-380, 1997.

2. Look AT, Hayes FA, Shuster 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.

3. O’Reilly R: NCCN Pediatric neuroblastoma guidelines. Oncology 10:813-822, 1996.

4. Cohn SL, Cook AT, Joshi VV, et al: Lack of correlation of N-myc gene amplification with prognosis in localized neuroblastoma: A Pediatric Oncology Group study. Cancer Res 55:721-726, 1995.

5. Hasse GM, Wong KY, DeLorimier AA, et al: Improvement in survival after excision of primary tumor in stage III neuroblastoma. J Pediatr Surg 24:194-200, 1989.

6. LaQuaglia MP, Kushner BH, Heller G, et al: Stage 4 neuroblastoma diagnosed at more than one year of age: Gross total resection and clinical outcome. J Pediatr Surg 29:1162-1166, 1994.

7. Kiely EM: The surgical challenge of neuroblastoma. J Pediatr Surg 29:128-133, 1994.

8. Berthold F, Utsch S, Holsehneider AM: The impact of preoperative chemotherapy on resectability of primary tumor and complication rate in metastatic neuroblastoma. Z Kinderchir 44:21-24, 1989.

9. Hasse GM, O’Leary ML, Ramsay NK, et al: Aggressive surgery combined with intensive chemotherapy improves survival in poor-risk neuroblastoma. J Pediatr Surg 26:1119-1124, 1991.

10. Nakagawa A, Ikeda K, Yokoyama T, et al: Surgical aspects of N-myc oncogene amplification of neuroblastoma. Surgery 104:34-40, 1988.