Neuroblastoma, a malignancy of neural crest origin, is the commonest solid tumor in young children. With a median age at diagnosis of two and one-half years, 90% of patients will be diagnosed in the first five years of life. A characteristic feature of this malignancy is early dissemination, with 70% of patients having tumor spread beyond the primary site at the time of diagnosis. The sites of dissemination are to lymph nodes, bone marrow, bone, liver, and skin.
The site of origin of neuroblastoma may be in the adrenal gland or in any ganglia of the sympathetic nervous system. Because of this and the high frequency of early metastases in this disease, there are many presenting signs and symptoms. The asymptomatic child may be found to have an abdominal or pelvic mass on examination for a minor illness unrelated to the tumor, emphasizing the importance of a good abdominal exam whenever a child is seen by a physician. Thoracic tumors are often found when a chest x-ray is taken for suspected pulmonary disease (Figures 1 and 2). A unilateral neck mass (Figure 3), especially in the absence of upper respiratory symptoms or in association with a Horner's syndrome may be the presenting feature of this disease. The most frequent symptoms of metastatic disease to bones and marrow are pain, limp and refusal to walk.
It is rare, even with almost total marrow replacement by tumor (Figure 4), for children with neuroblastoma to have any significant abnormality of the blood count; a normal hemoglobin, white blood count and platelet count does not indicate absence of significant marrow disease. Many children present with nonspecific symptoms of anorexia, weight loss, low grade fever, abdominal discomfort and generalized aches and pains. The physician must maintain a high degree of suspicion of malignancy in any child who has these complaints recurrently or for a longer time than one would expect for a viral syndrome, etc. Other signs of neuroblastoma include periorbital edema and ecchymosis (Figure 5), skin nodules (Figure 6), supraclavicular node enlargement, skull masses, signs of spinal cord compression and hepatomegaly (Figure 7). Rarely a child with this disease may present with hypertension, diarrhea oran opsiclonusmyoclonus syndrome.
The only independent clinical prognostic factors in neuroblastoma are the stage or extent of disease at diagnosis and the age of the child at diagnosis. For all stages of disease beyond localized completely excised tumors, the infant less than one year of age has a significantly better remission rate and survival than the older child of equivalent stage. There is no survival advantage for the child who is between one and two years of age.
As with most malignancies, stage of disease at diagnosis is the single most important prognostic factor in neuroblastoma. By the clinical staging classification of Evans et al,' approximately 40% of children will appear to have localized, non-disseminated disease. Evaluation of these patients using a surgical-pathological staging system, however, reveals that a proportion of these children have dissemination of tumor to regional lymph nodes.2'3 The presence of tumor in regional lymph nodes markedly alters the prognosis in these patients, and they have a clinical course much the same as the child with more widely disseminated tumor (Figure 8). Studies, showing the prognostic significance of regional node involvement with tumor, indicate that pre-therapy evaluation of lymph nodes is essential for therapy planning and for meaningful evaluation of any therapy delivered to the child whose disease clinically appears to be localized.
Figure 1. Right paraspinal neuroblastoma.
Figure 2. Chest x-ray of patient presenting with a right Horner's syndrome showing a right upper thoracic neuroblastoma.
Figura a Child with cervical primary neuroblastoma with bilateral proptosis and periorbital edema.
Pre-therapy Evaluation Prior to any surgical intervention, any child suspected of having neuroblastoma should have a complete evaluation for metastatic disease. This should include a bone marrow aspiration, bone scan, skeletal survey, liver scan or computerized axial tomography (CT) of the liver. Urinary catecholamine levels should be obtained, realizing that not all patients with neuroblastoma will have elevated values.
Surgery The role of the surgeon in the current management of neuroblastoma is an expanded one. In the child in whom a diagnosis cannot be made on the basis of elevated urinary catecholamines and the finding of characteristic bone marrow tumor, surgical-pathological confirmation of the diagnosis must be obtained.
For the child who presents with clinically localized tumor the surgeon's role is twofold. Lymph node biopsies, separate from the primary tumor, plus a liver biopsy if the tumor is in the abdomen, are essential for correct staging of the patient. An attempt should be made to excise completely the primary tumor if this can be done without undue risk to the child. If the primary tumor is obviously unresectable, there is no evidence that partial resection, as a primary approach, is of any advantage to the patient.
Figure 4. Neuroblastoma metastatic to bone marrow (Wright's stain, magnification 80Ox).
Figure 5. Subtle bilateral sub-orbital ecchymosis which was a presenting sign of disseminated neuroblastoma.
Figure 6. Subcutaneous nodule of neuroblastoma.
Figure 7. Liver scan demonstrating extensive infiltration by neuroblastoma.
The surgical role in neuroblastoma is not only at the time of diagnosis. For patients with unresectable localized tumors at diagnosis, interval therapy with radiation or chemotherapy frequently results in significant regression of tumor so that a delayed "second look" procedure for complete excision may be curative.
Radiation Therapy The role for conventional radiation therapy in the primary treatment of neuroblastoma is unclear. It is generally accepted that the child with complete gross excision of a localized primary tumor, even when there is microscopic residual in the tumor bed, with few exceptions is cured by the surgical procedure. This is true regardless of age, site of primary or size of primary. Of the 33 children with grossly excised primary tumors depicted in Figure 8, 1 5 of 1 6 diagnosed since 1 975 survive disease-free, although they received no therapy beyond their surgical procedure.
The use of radiation therapy for the child with localized but unresectable tumor is more controversial. We have taken the approach of giving only chemotherapy in a regimen consisting of sequential cyclophosphamide and adriamycin, to induce tumor regression. After four months of therapy, a "second look" surgical procedure is performed.2 Of 16 patients treated since 1974 with this approach, 14 survive disease-free, 11 of whom never ha ve received any radiation therapy. Since the results suggest that the majority of these children will survive their disease if they are given only chemotherapy, one must question the routine exposure of these young children to the long-term hazards of radiation therapy.4 Forchildren with disseminated tumor, there is no evidence that radiotherapy improves survival. One study utilizing total body sequential segmental irradiation in combination with chemotherapy did not improve disease-free survival, with most patients showing recurrence of tumor, even before all radiation therapy was delivered.5
There is no question that radiation therapy delivered to painful or disfiguring lesions can have a beneficial but palliative effect in this tumor.
Chemotherapy For the majority of patients presenting with neuroblastoma, surgery and radiotherapy will be of little benefit and chemotherapy control of disseminated disease is needed. Of the many agents tested in patients with neuroblastoma only six, as single agents, have resulted in a significant complete and partial remission rate of greater than 20%. There are cyclophosphamide, adriamycin, cis-dichlorodiamine platinum (CDDP), the epipodophyllotoxin VM26, vincristine, and peptichemio. Various combinations of these agents and others of lower single agent efficacy have been used in multiple trials in chemotherapy of this disease.
In evaluating chemotherapy trials in neuroblastoma, one must take into account the age of the child and the stage of disease when therapy is initiated. It is also unrealistic at this time to use survival as an end point of the success or failure of therapy in this disease. A more realistic evaluation is the ability of the therapy program to induce complete remission, since until complete remission is attained by a high proportion of patients a meaningful improvement in disease-free survival will not occur. Figure 9 graphically demonstrates the relationship of duration of survival to the response to induction chemotherapy and the importance of increasing the complete remission rate.
Recent studies from St. Jude Children's Research Hospital would indicate that there are two groups of patients in whom the complete remission rate and diseasefree survival following chemotherapy can be expected to be greater than 75%. One group is those children described above who have localized but unresectable disease.2 The second group consists of infants less than one year of age with disseminated disease who were treated with sequential cyclophosphamide and adriamycin.7 Of 24 infants with non-stage IVS disease, treated with this therapy, 1 9 attained complete remission while two others attained remission when CDDP and VM26 were added to the chemotherapy regimen.8 Of 21 attaining complete remission, only two have developed recurrent tumor.
Figure 8. Survival of patients with neuroblastoma in relation to surgical-pathological stage of disease at diagnosis. ( _____) completely excised localized (negative nodes) N =33. ( _____ ) unresectable localized (negative nodes) N=33, ( _____ ) regional lymph nodes positive for tumor N=36, and ______ ) widely disseminated disease N=145.
Figure 9. Survival of patients with neuroblastoma with disseminated disease according to remission status following four months of induction chemotherapy.
However, the largest group of children with neuroblastoma are those over the age of one year who present with metastatic disease. For this group of children the proportion surviving long-term disease-free has changed little. However, there has been improvement in the complete remission rate, which is the first essential step in improving survival. The first combination of drugs used to treat these children was cyclophosphamide and vincristine. The overall complete remission rate with this combination was about 21%.9'10
The Children's Cancer Study Group evaluated the effectiveness of adding Dacarbazine (DTIC) and the combination of DTIC plus adriamycin to vincristine and cyclophosphamide. They reported 38% and 43% complete remission rates with these two chemotherapy regimens.11 Papaverine and trifluoromethyl-deoxyuridine, which cause maturation of neuroblasts in vitro, were added to vincristine and cyclophosphamide in studies reported by the Southwest Oncology Group and the Memorial Sloan-Kettering Cancer Center. This regimen resulted in a complete remission rate of 52% in previously untreated patients.12'13 However, this improved complete remission rate was accompanied by prohibitive toxicity in many patients. The sequential scheduling of low dose cyclophosphamide and adriamycin reported by St. Jude Children's Research Hospital also produced a complete remission rate of 52%, but with minimal associated toxicity.7'14
Since phase II trials with CDDP and VM26 indicated anti -tumor effect in neuroblastoma, combination therapy with these agents was tested. The sequential scheduling of CDDP, followed in 48 hours by VM26, was tested in 22 previously treated patients; six of the 22 patients attained complete remission and nine attained partial responses.8 The effectiveness of this combination is now being evaluated in previously untreated patients by the Pediatric Oncology Group.
1. Evans A, D'Angio G, Randolf JA: Proposed staging for children with neuroblastoma. Cancer 1971; 27:374-378.
2. Hayes FA, Green AA, Hustu HO, et al: Surgical-pathological staging of neuroblastoma: The prognostic significance of regional lymph node metastases. J Pediatr, to be published.
3. Ntnane J, Pritchard J. Morris-Jones P, et al: Adverse prognostic significance of lymph node involvement in state II neuroblastoma. Arch Dis Child 1982: 57:438-442.
4. Meadows AT, D'Angio G, Evans A.etal: Oncogenesis and other late effects of cancer treatment. Radiology 1975: 114:175.
5. Green AA, Hustu HO, Palmer R, et al: Total-body sequential segmental irradiation and combination chemotherapy for children with disseminated neuroblastoma. Cancer 1976; 38:2250-2257.
6. Carli M, Green AA, Hayes FA, et al: Therapeutic efficacy of single drugsfor childhood neuroblastoma; a review. International Society of Pediatric Oncology, Marseilles, France, 1981, in Raybaud C, Clement R, LcbreuilG, et al (eds): International Congress Series No 570. New York, ExcerptaMedka. 1981, pp 141-150.
7. Green AA, Hayes FA, Hustu HO: Sequential cyclophosphamide and doxyrubicin for induction of complete remissions in children with disseminated neuroblastoma. Cancer 1981; 48:2310-2317.
8. Hayes FA, Green AA, Casper J, et al: Clinical evaluation of sequentially scheduled cisplatin and VM26 in neuroblastoma; response and toxicity. Cancer 1981; 48:1715-1718.
9. Pinkel D, Pratt C, Holt on C, et al: Survival of children with neuroblastoma treated with combination chemotherapy. / Pediatr 1968; 73:928-931.
10. Sawitsky A Vincristine and cyclophosphamide therapy in generalized neuroblastoma; a collaborative study. Am J Dis Child 1970; 119:308-313.
11. Finkelstein J, Klemperer M, Evans A, et al: Multiagent chemotherapy for children with metastatic neuroblastoma; a report from Children's Cancer Study Group. Med Pediatr Oncol 1979; 6:179-188.
12. Hclson L, Helson C, Peterson R. et al: A rationale for the treatment of metastatic neuroblastoma. Journal of the National Cancer Institute 1976; 57:727-729.
13. Nttschke R, Cangir A, Crist W. et al: Intensive chemotherapy for metastatic neuroblastoma; a Southwest Oncology Group Study. Med Pediatr Oncol 1980;8:281-288.
14. Hayes FA, Green AA. Mauer AM: Correlation of cell kinetics and clinical response to chemotherapy in disseminated neuroblastoma. Cancer Res 1977; 37:3766-3770.