- Email: [email protected]
Excellent Local Control From Radiation Therapy for High-Risk Neuroblastoma
Int. J. Radiation Oncology Biol. Phys., Vol. 74, No. 5, pp. 1549–1554, 2009 Copyright Ó 2009 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/09/$–see front matter
doi:10.1016/j.ijrobp.2008.10.069
CLINICAL INVESTIGATION
Pediatrics
EXCELLENT LOCAL CONTROL FROM RADIATION THERAPY FOR HIGH-RISK NEUROBLASTOMA HEATHER G. GATCOMBE, M.D.,* R. B. MARCUS, JR., M.D.,x HOWARD M. KATZENSTEIN, M.D.,y MOURAD TIGHIOUART, PH.D.,z AND NATIA ESIASHVILI, M.D.* * Department of Radiation Oncology, y Children’s Healthcare of Atlanta, z Biostatistics and Informatics at Winship Cancer Institute, Emory University School of Medicine, Atlanta, GA; and x Department of Radiation Oncology, University of Florida, Gainesville, FL Purpose: Local recurrence has been demonstrated in previous studies to be one of the obstacles to cure in neuroblastoma. Radiation therapy indications, optimal dose, and technique are still evolving. Here we report our experience of high-risk neuroblastoma patients who received local radiation therapy as part of their cancer management. Methods and Materials: We conducted a retrospective study of 34 high-risk neuroblastoma patients who received radiation therapy to local sites of disease from March 2001 until February 2007 at our institution as part of their multimodality therapy. Results: At a median follow-up of 33.6 months, 6 patients died of disease, 7 patients were alive with disease, and 21 patients were in clinical remission. Eleven patients relapsed, all distantly. Two patients failed locally in addition to distant sites. Both of these patients had persistent gross disease after induction chemotherapy and surgery. Our 3-year local control, event-free survival, overall survival were 94%, 66%, and 86%, respectively. Conclusion: Patients with high-risk neuroblastoma in our series achieved excellent local control. Doses of 21–24 Gy to the primary tumor site appear to be adequate for local control for patients in the setting of minimal residual disease after induction chemotherapy and surgery. Patients with significant residual disease may benefit from radiation dose escalation, and this should be evaluated in a prospective clinical trial. Ó 2009 Elsevier Inc. High risk, Neuroblastoma, Local control, Radiation.
remain high in larger cooperative group studies. The Children’s Oncology Group (COG) in its current high-risk protocol will evaluate dose escalation in patients with gross disease to a dose of 36 Gy, whereas patients with microscopic residual disease will continue to receive 21.6 Gy. It is important to identify the subset of patients with a higher risk of local failure and determine whether they would benefit from dose escalation. The question of dose response on local tumor control is especially crucial in this young patient population in which radiation late effects are of high concern. Herein, we report our institutional experience in treating high-risk neuroblastoma with multimodality therapy incorporating local external beam radiation therapy (EBRT) with specific focus on local disease control. In addition to known patient and tumor-related risk factors, we analyzed whether radiation therapy dose and extent of surgical resection affected local disease control.
INTRODUCTION Neuroblastoma is the most common extracranial solid tumor of childhood. Approximately 50% of newly diagnosed patients have high-risk disease, and despite aggressive multimodality treatment incorporating induction chemotherapy, maximum tumor resection, myeloablative treatments, and maintenance therapy with cis-retinoic acid, and monoclonal antibody, these children do poorly with long-term survival rates of less than 40% (1–6). Although distant relapse is the main obstacle to cure in these patients, local failure remains a significant problem. The addition of radiation therapy to the multimodality approach of these high-risk patients has resulted in improvement in local control as supported from a large cooperative group study and several institutional reports (7–9). The optimal radiation therapy indications, technique, and dose are still evolving. Although there have been no randomized trials of radiation therapy in patients with high-risk disease, there was evidence of a dose response in the Children’s Cancer Group (CCG-3891) trial (7). Local failure rates, however,
METHODS AND MATERIALS After receiving internal review board approval, we conducted a retrospective chart review of all high-risk neuroblastoma patients
Reprint requests to: Natia Esiashvili, M.D., Department of Radiation Oncology, Emory School of Medicine, 1365 Clifton Road, NE, Atlanta, GA 30322. Tel: (404) 778-5782; FAX: (404) 7784139; E-mail: [email protected]
Conflict of interest: none. Received Aug 5, 2008, and in revised form Oct 15, 2008. Accepted for publication Oct 16, 2008. 1549
1550
I. J. Radiation Oncology d Biology d Physics
who received radiation therapy in our department from 2001 to 2007. All 34 patients were International Neuroblastoma Staging System (INSS) Stage 3 and 4 high-risk neuroblastoma as defined by the COG Risk Stratification Schema. All patients had histologic confirmation of their disease with either biopsy of their primary or metastatic tumor site or the presence of neuroblastoma in the bone marrow with elevated urinary catecholamines. The majority of patients were enrolled in the COG neuroblastoma biology study to determine biologic variables. Staging workup was performed with imaging including computed tomography (CT) of the chest and abdomen, 123I-meta-iodobenzylguanidine (MIBG) scan, technetium-99 bone scan, and bilateral bone marrow aspirates and biopsy. Patients were staged according to INSS criteria and assigned to the high-risk group according to COG risk-stratification scheme (Stage 4 disease in patients older than 1 year, MYCN-amplified tumors in patients with Stage 2, 3, or 4S disease or with Stage 4 disease and younger than 1 year, or unfavorable histology in patients older than 1 year with Stage 3 disease) (10). The majority of patients were treated either on or according to recent COG protocols ANBLOOP1 and ANBL02P1, including induction chemotherapy with cyclophosphamide, doxorubicin, vincristine, cisplatin, etoposide, and topotecan. Patient response was evaluated by bone marrow, bone scan, MIBG, and CT. Surgery was attempted on all patients unless they had a complete or near-complete response to chemotherapy. The goal of surgery was a gross total resection with preservation of vital organs. Near or gross total resection (NTR/GTR) was defined as > 95% surgical resection. Patients then underwent either single or tandem autologous stem-cell transplantation following high-dose chemotherapy. After recovery from myelosuppression, patients were treated with EBRT to primary tumor sites regardless of the extent of surgical resection and metastatic sites as indicated on the basis of their chemotherapy response. CT planning was used to define the treatment volume with customized blocking. EBRT was delivered by linear accelerator, 6-MV energy. The gross tumor volume (GTV) consisted of the postchemotherapy, presurgical disease. For abdominal primaries, the clinical tumor volume (CTV) incorporated the para-aortic lymph nodes in addition to GTV. The planning target volume (PTV) was created by expanding the CTV 1–2 cm, taking into account anatomic constraints. In most cases, opposed anterior and posterior fields were used. An intensity-modulated radiation therapy (IMRT) plan was performed for six patients with similar guidelines for targeting primary tumor and regional lymph nodes in the PTV and appropriate dose constraints to protect normal tissues. Daily fraction size was 1.5, 1.8, or 2 Gy daily and, in one case, 1.2 Gy twice daily. Six patients were simultaneously treated to metastatic sites of disease for persistent metastatic disease as determined from postinduction MIBG and bone-scan positivity. The decision to irradiate sites of distant metastatic disease was at the discretion of the treating physician and parental choice. Three patients received total body irradiation (TBI) for transplant conditioning consisting of 1.5 Gy twice daily for 4 days with at least a 6-hour separation between morning and afternoon fractions. Lung blocks were used to decrease the lung dose. After completion of radiation therapy, patients received adjuvant cis-retinoic acid for six cycles as part of their maintenance therapy for treating minimal residual disease. Patients were followed after completion of treatment in the pediatric hematology/oncology and radiation oncology clinics with clinical examinations and appropriate laboratory and imaging testing to evaluate disease control. All of the disease relapses were documented with imaging and correlated with known adverse disease factors (MYCN amplification, Shimada histopathology), as well as with response to induction che-
Volume 74, Number 5, 2009
motherapy, absence or presence of residual disease in the primary site after surgery, number of autologous bone marrow transplants, and radiotherapy parameters. Diagnostic imaging documenting relapse was compared with RT fields and dosimetry to determine failure inside or outside of treatment fields.
Statistical analysis Relapse-free survival was calculated from the date of diagnosis to date of relapse or progression or to date of last follow-up if patients remained in remission. Local relapse-free survival was calculated from date of diagnosis to date of local failure at primary tumor site as seen on CT, MIBG, or bone scan. Overall survival was calculated from date of diagnosis to date of last follow-up or date of death. Kaplan-Meier statistics were used to calculate actuarial disease-free, overall, and local relapse-free survival curves.
RESULTS From March 2001 until February 2007, we treated 34 (15 male, 19 female) children with high-risk neuroblastoma with local radiation to their primary site of disease. Patient characteristics are summarized in Table 1. Median age at diagnosis was 3.4 years (range, 10 months–20 years). Five patients had INSS Stage 3 and 29 Stage 4 neuroblastoma. The primary tumor site was abdominal in 27 patients, thoracic in 3 patients, pelvic in 2 patients, and combined in 2 patients. MYCN was amplified in 12 patients, nonamplified in 16, and unknown in 6 patients. Shimada classification was unfavorable in 26, favorable in 3, and unknown in 5. Following induction chemotherapy, 24 patients had persistent disease, and 10 demonstrated a complete response as determined by postinduction scans. Surgery was performed after induction chemotherapy and before transplantation. Eighteen patients had a gross or near total resection (GTR/ NTR), 12 achieved a subtotal resection (STR), 1 patient had a biopsy alone, and 3 did not undergo surgery. The patient who had a biopsy alone had unresectable disease due to involvement of the aorta, inferior vena cava, and superior mesenteric artery. The 3 patients who did not undergo a surgical resection had a complete or near complete response to induction chemotherapy on the basis of radiographic and functional imaging. The remaining 7 patients who had a complete response following chemotherapy based on functional imaging (MIBG and bone scan) still had persistent small-volume residual disease on CT scan for which they were taken to surgery. Following resection, 6 patients underwent a single transplant and 28 underwent tandem transplant. The reasons for undergoing a single transplant only were complications from surgery or chemotherapy (2 patients), Stage 3 disease (1 patient), inadequate stem-cell counts for tandem transplant (1 patient), and parental refusal (1 patient); another patient failed to clear his bone marrow and went on to receive MIBG therapy followed by a single transplant. Three patients received total body irradiation (TBI) in preparation for transplantation. These patients were treated according to bone marrow transplant protocols adopted before March 2002 and received 12 Gy at 1.5 Gy twice daily. The remaining
RT for high-risk neuroblastoma d H. G. GATCOMBE et al.
Table 1. Patient Characteristics (34 patients) Age Median Range Sex (n) Male Female INSS stage 3 4 Primary site Abdomen Thorax Pelvis Combined Metastatic sites Skeletal Bone marrow Distant lymph node Orbit Lung Response to induction chemo (based on MIBG and bone scan) CR Persistent disease Resection extent GTR/NTR STR Biopsy only No surgery Single transplant Tandem transplant
Table 2. Radiation Dose Radiation dose (Gy)
3.4 years 10 months–20 years 15 19 5 29
1551
Primary site dose (excluding TBI ; n = 34)* 12–18 19–24 25–30.6 Metastatic sites (n = 6) 21–22 22–23 23–24
n (% patients) 3 (9) 27 (79) 4 (2) 4 (67) 1 (17) 1 (17)
27 3 2 2
Abbreviation: TBI = total body irradiation. * Three patients received TBI to 12 Gy (1.5 Gy twice daily) in addition to primary site radiation doses of 12, 18, and 18 Gy.
28 23 9 1 2
3-year overall survival was 86%. Our 3-year local control was 94% (Fig. 2).
PATTERNS OF RELAPSE 10 (29%) 24 (71%) 18 12 1 3 6 28
Abbreviations: CR = complete response; GTR = gross total resection; INSS = International Neuroblastoma Staging System; MIBG = 123I-meta-iodobenzylguanidine; NTR = near total resection; STR = subtotal resection.
31 patients received high-dose chemotherapy for myeloablative conditioning. The median time from last transplant until the start of radiation was 61.5 days (range, 38–136 days). All 34 patients received radiation to the primary disease site with the majority treated to the abdomen (27 abdominal, 3 thoracic, 2 pelvic, and 2 combined abdominal and thoracic fields). The median radiation dose was 22 Gy (range, 18.0– 30.6), which includes the cumulative dose for the 3 patients who additionally received TBI (Table 2). The most common fraction schemes used were 21.6 Gy at 1.8 Gy per day (10 patients) and 21 Gy at 1.5 Gy per day (8 patients). During radiotherapy to their primary site of disease, 6 patients received concurrent RT to persistent sites of metastatic sites of disease (dose range, 21 Gy at 1.5 Gy/fraction to 30.6 Gy at 1.8 Gy/ fraction) as seen on postinduction scans (Table 3). The majority of patients required general anesthesia because of their young age. All but 1 patient received adjuvant cis-retinoic acid as part of their maintenance therapy. The median follow-up was 33.6 months (range, 0.5–70.5 months). At the time of analysis, 21 patients had no evidence of disease (NED), 7 were alive with disease (AWD), and 6 were dead of disease (DOD). Eleven of the 34 patients relapsed, all distantly and 2 with concurrent local failure. The 3-year event-free survival (EFS) was 66% (Fig. 1), and
Sites of distant relapse included skeletal (8 patients), bone marrow (5 patients), lung (4 patients), liver (1 patient), leptomeningeal (1 patient), and face/orbit (1 patient). The majority of patients who relapsed in bone and bone marrow had disease present there at initial presentation. The majority of patients with lung, liver, and leptomeningeal relapse did not present with disease at these locations at initial diagnosis. Of the 11 patients who relapsed, all but one were INSS Stage 4. MYCN amplification and histology type in our series was not strongly associated with the risk of relapse. Seven of the 11 relapsed patients demonstrated distant persistent disease on MIBG or bone scan following induction chemotherapy. One of the 11 did not have a bone or MIBG scan postinduction, but CT demonstrated a residual primary tumor mass measuring 6 8cm. Four of the 11 patients had undergone a GTR/NTR, 6 had a STR, and 1 had a biopsy only. Extent of resection and response to induction chemotherapy were found to have no statistically significant influence on EFS. The majority of relapsed patients did not receive radiotherapy to metastatic sites. Of 6 patients treated to metastatic lesions, only 1 suffered progression inside the radiation field. When examined by the number of transplantations completed, patients who received tandem stem-cell rescue seemed to have a better outcome than patients receiving only a single stem-cell rescue (p = 0.043, Fig. 3). The median age of patients undergoing a single or tandem stem-cell rescue was significantly different (6.3 and 2.9 years, respectively). Moreover, the younger patients were more likely to undergo tandem stem-cell rescue and overall had better outcome (HR 1.18, p = 0.0256). After age was taken into account, the number of stem cell rescues were no longer significant. Both patients with local failure were INSS Stage 4 with unfavorable histology and MYCN amplification. They both presented with bone marrow and skeletal metastases at diagnosis and had persistent disease seen after induction chemotherapy (one locally and the other distantly). Both underwent
1552
I. J. Radiation Oncology d Biology d Physics
Volume 74, Number 5, 2009
Table 3. Metastatic disease location and local control of irradiated metastatic lesions
Bone Distant lymph nodes Other
No. of Patients with metastatic disease at Presentation*
No. of Patients with metastatic disease after induction chemotherapy*
No. of metastatic sites irradiated
Local control of irradiated metastatic sites
28 9 3
13 0 1
3 2y 1
2 2 1
* By 123I-meta-iodobenzylguanidine and bone scan. y Both patients had persistent enlarged supraclavicular lymph nodes on CT imaging after induction chemotherapy.
1.00
Event Free Survival
0.75
3-year Event-Free Survival 66%
0.50
High-risk neuroblastoma patients continue to present many challenges for treatment. These children often present with widely metastatic disease and large, bulky, invasive tumors involving critical vasculature and a gross resection is often difficult to obtain. Despite aggressive multimodality therapy including intensification of systemic therapy, local relapse remains an obstacle to cure in high-risk neuroblastoma patients. Adjuvant radiation therapy with EBRT, intraoperative radiotherapy (IORT), or both has been incorporated into high-risk neuroblastoma protocols to improve local control, but the optimal dose to deliver to these patients is unknown (7, 9, 11–13). Careful consideration of radiation dose and volume is important given the young age of these patients and the abutting critical normal structures. What is the optimal dose for these young patients? The addition of local radiation therapy to aggressive systemic therapy and tumor resection has resulted in good local control rates in institutional studies. In our study group, the local control at the primary tumor site was 94% at 3 years. Our results are consistent with previous reports of high-risk patients treated with aggressive multimodality therapy including double or triple transplantation and incorporating local radiation
0.25
DISCUSSION
therapy (8, 9, 12, 14). Marcus et al. (9) reported an actuarial probability of local control of 97% with a median follow-up of 19.4 months in patients with Stage 4 or high-risk Stage 3 neuroblastoma treated with induction chemotherapy, tandem transplant, surgery, and local radiotherapy (12 Gy TBI plus local EBRT dose 10.5–18 Gy). Bradfield et al. (14) reported a local failure rate of 7% (95% confidence interval, 0%–14%) at 2 years after delivering 21 Gy at 1.5 Gy a day. Kushner et al. (8) reported primary site failure rates of 4.1% 3.0% at 12 months and 10.1% 5.3% at 36 months from the start of radiation; their protocol consisted of 21 Gy in 1.5-Gy fractions twice daily to all high-risk neuroblastoma patients, regardless of resection status. The Chicago Pilot II Protocol achieved a local control rate of 97% in a cohort of 30 highrisk patients with abdominal primaries who received 24 Gy and triple-tandem high-dose therapy (12). Taken together, the data may suggest that doses of 21–24 Gy are adequate for local control for the majority of patients with high-risk neuroblastoma, but these results were hard to reproduce in the cooperative group setting. In our study, only six patients were treated with IMRT, and the majority received conventional radiotherapy. Evaluation of various radiation delivery techniques have been previously reported from our institution in an effort to limit the radiation dose to normal tissues and decrease potential toxicities (15). IMRT was not found to be beneficial in patients with lateralized tumors because it delivered a higher radiation dose to the spleen, liver, and stomach and contralateral kidney and was used less frequently in more recently treated patients.
0.00
a STR with gross residual disease left at the primary site. The first patient underwent tandem stem-cell rescue, with radiation starting 62 days after his second rescue. He had 1.3 cm of residual disease as documented on CT before initiation of radiotherapy. The second patient had a biopsy only for a large tumor encasing the aorta, inferior vena cava, and superior mesenteric artery measuring 2.8 8.0 6.0 cm on CT. He received tandem stem-cell rescue with local radiation starting 108 days after his last rescue. The first patient received 22.5 Gy, and the second patient received 12Gy TBI and then 12 Gy local radiation to residual disease. Both patients suffered from both local-regional and distant failures simultaneously and died of disease. Of the 6 patients who also received concurrent radiation to metastatic sites of disease, there was only one treatment failure inside the radiotherapy field. A patient who received 21.6 Gy to a left femur metastasis recurred at this site in addition to lung and other skeletal sites 38 months after completing radiation. An additional patient continues to exhibit a persistent abnormality on bone scan after receiving 23.4 Gy to the calvarium, but this area at last follow-up has remained stable without evidence of disease progression or relapse.
0
20
40
Time (months)
Fig. 1. Event-free survival.
60
80
RT for high-risk neuroblastoma d H. G. GATCOMBE et al.
1.00
Local Control
1553
Relapse Free Survival Kaplan-Meier Curve
100
3-year Local Control 94%
Numb Transp 1 2
0.75
90
0.25
0.50
Percent
80 70 60 50
0.00
40 0 0
20
40
60
80
Time (months)
10
20
30
40
50
60
70
80
Time to Relapse (In months)
Fig. 2. Local control.
Fig. 3. Event-free survival by number of transplants.
The optimal radiation target volume is still disputable. In our series, consistent with other institutional protocols, we included the postinduction chemotherapy and presurgical tumor volume plus a 2-cm margin. However, we also elected to cover the first-echelon draining lymph nodes for all patients with abdominal primaries. Some groups argue against including the uninvolved draining lymph nodes within the radiation and field, and it was not allowed on the preceding COG protocols. It will be important to know whether this had any impact on disease control—specifically on the incidence of regional nodal failures, which are sometimes hard to distinguish anatomically from primary site recurrence. Of our 11 patients who relapsed, 4 received a GTR/NTR, 6 had an STR, and 1 patient had a biopsy alone. There have been conflicting reports of whether extent of resection improves outcome. Browne et al. (12) found that although patients with a GTR had lower overall recurrence rates, there was no significant difference in the local recurrence rate. Analysis of the CCG-3891 trial demonstrated no improvement in event-free or overall survival by achieving a complete resection (16). Alternatively, La Quaglia et al. (17) found a 50% probability of local progression in unresected patients compared with 10% in patients undergoing GTR (p <.01). Their overall survival rate in resected patients was 50% compared with 11% in unresected patients (p < .01), and they recommended that GTR should be part of the management of Stage 4 neuroblastoma in patients older than 1 year (17). Matthay et al. (18) demonstrated a survival improvement for Stage 3 tumors with unfavorable biology with complete or near complete resection. Gillis et al. (19) also found a significant difference in patients undergoing GTR or STR who had been treated with IORT (19). Although the extent of surgical resection on EFS did not reach statistical significance, we emphasize the importance of achieving maximal surgical resection, given that, in our cohort, two patients who had significant residual disease at the primary site experienced local failure. These children went on to receive doses of 22.5 Gy (1.5 Gy daily) and 24 Gy (12 Gy TBI and 12 Gy to local site). Would patients with persistent gross disease benefit
from dose escalation? Further investigation is necessary to answer this question. An upcoming COG study plans on delivering doses of 21.6 Gy after GTR and 36 Gy to residual disease (20) . Aggressive initial surgical resection or second-look surgery is another consideration to improve local control, but, as with radiotherapy, it would require a careful balance between benefit and the surgical morbidity risk, which remains high (21, 22). One of the obstacles in evaluating the extent of surgery has been assessing the completeness of surgical resection. Von Allmen et al. (21) has reported a 48% discrepancy between the surgeons’ assessment of residual tumor and the imaging findings. Reliable methods for detecting viable residual tumor vs. mature scar tissue are essential for accurate selection of appropriate candidates for radiotherapy dose escalation. Our data show that even after STR, local control was possible in the majority of patients. The amount of residual tumor measured volumetrically may also have some significance and warrants careful evaluation in a cooperative group setting. Radiation therapy indications for neuroblastoma metastatic sites are also not standard. Most institutional reports, in fact, do not provide sufficient information on treating metastatic sites. The cooperative group studies including those recently completed and currently open are recommending radiation to persistent sites of metastatic disease as seen on MIBG and bone scans. In our patient population, most patients were treated according to COG criteria, and only one of our patients failed at an irradiated metastatic site (an additional patient continues to have a persistent abnormality on bone scan). The majority of patients who relapsed with skeletal metastasis had disease present at these locations at the time of initial diagnosis and could possibly have benefited from radiation to these sites regardless of responsiveness to induction chemotherapy. This question still warrants careful evaluation in prospective fashion. It is well-established that intensive chemotherapy including myeloablative high-dose chemotherapy with autologous stem-cell rescue adds to disease control. The number of rescues may have potential influence on disease outcome
1554
I. J. Radiation Oncology d Biology d Physics
(23–25). In our analysis, patients who underwent tandem stem cell rescue had a better outcome than those who received a single rescue. However, after this variable was adjusted for age, the number of transplants were no longer significant. Only two patients did not receive a second rescue because of complications from chemotherapy or surgery, and it is thus difficult to ascertain whether younger patients were able to tolerate the second rescue better or they have a better prognosis regardless of therapy type. In conclusion, radiation therapy continues to play an important role in the local control of high-risk neuroblastoma.
Volume 74, Number 5, 2009
Our institution has achieved excellent local control in our high-risk neuroblastoma patients. Doses of 21–24 Gy to the primary tumor site appear to be adequate for local control for patients in the setting of minimal residual disease. Although in our series most patients with residual disease still achieved good local control, patients with significant residual tumor may benefit from radiation dose escalation. A prospective randomized trial is necessary to determine whether patients with persistent gross disease after induction chemotherapy and surgery would benefit from further dose escalation.
REFERENCES 1. De Bernardi B, Nicolas B, Boni L, et al. Disseminated neuroblastoma in children older than one year at diagnosis: Comparable results with three consecutive high-dose protocols adopted by the Italian Co-Operative Group for Neuroblastoma. J Clin Oncol 2003;21:1592–1601. 2. Matthay KK, Villablanca JG, Seeger RC, et al. Treatment of high-risk neuroblastoma with intensive chemotherapy, radiotherapy, autologous bone marrow transplantation, and 13-cisretinoic acid. Children’s Cancer Group. N Engl J Med 1999; 341:1165–1173. 3. Matthay KK. Impact of myeloablative therapy with bone marrow transplantation in advanced neuroblastoma. Bone Marrow Transplant 1996;18(Suppl. 3):S21–S24. 4. Kushner BH, LaQuaglia MP, Bonilla MA, et al. Highly effective induction therapy for stage 4 neuroblastoma in children over 1 year of age. J Clin Oncol 1994;12:2607–2613. 5. Cheung NK, Kushner BH, Cheung IY, et al. Anti-G(D2) antibody treatment of minimal residual stage 4 neuroblastoma diagnosed at more than 1 year of age. J Clin Oncol 1998;16: 3053–3060. 6. Stram DO, Matthay KK, O’Leary M, et al. Consolidation chemoradiotherapy and autologous bone marrow transplantation versus continued chemotherapy for metastatic neuroblastoma: a report of two concurrent Children’s Cancer Group studies. J Clin Oncol 1996;14:2417–2426. 7. Haas-Kogan DA, Swift PS, Selch M, et al. Impact of radiotherapy for high-risk neuroblastoma: A Children’s Cancer Group study. Int J Radiat Oncol Biol Phys 2003;56:28–39. 8. Kushner BH, Wolden S, LaQuaglia MP, et al. Hyperfractionated low-dose radiotherapy for high-risk neuroblastoma after intensive chemotherapy and surgery. J Clin Oncol 2001;19: 2821–2828. 9. Marcus KJ, Shamberger R, Litman H, et al. Primary tumor control in patients with Stage 3/4 unfavorable neuroblastoma treated with tandem double autologous stem cell transplants. J Pediatr Hematol Oncol 2003;25:934–940. 10. Brodeur GM, Kletzel M, Cohn SL, et al. Revisions of the international criteria for neuroblastoma diagnosis, staging, and response to treatment. J Clin Oncol 1993;11:1466–1477. 11. Haas-Kogan DA, Fisch BM, Wara WM, et al. Intraoperative radiation therapy for high-risk pediatric neuroblastoma. Int J Radiat Oncol Biol Phys 2000;47:985–992. 12. Browne M, Kletzel M, Cohn SL, et al. Excellent local tumor control regardless of extent of surgical resection after treatment
13. 14.
15. 16. 17. 18.
19. 20. 21.
22. 23. 24. 25.
on the Chicago Pilot II protocol for neuroblastoma. J Pediatr Surg 2006;41:271–276. Zachariou Z, Sieverts H, Eble MJ, et al. IORT (intraoperative radiotherapy) in neuroblastoma: Experience and first results. Eur J Pediatr Surg 2002;12:251–254. Bradfield SM, Douglas JG, Hawkins DS, et al. Fractionated low-dose radiotherapy after myeloablative stem cell transplantation for local control in patients with high-risk neuroblastoma. Cancer 2004;100:1268–1275. Paulino AC, Ferenci MS, Chiang KY, et al. Comparison of conventional to intensity modulated radiation therapy for abdominal neuroblastoma. Pediatr Blood Cancer 2006;46:739–744. Adkins ES, Sawin R, Gerbing RB, et al. Efficacy of complete resection for high-risk neuroblastoma: A Children’s Cancer Group study. J Pediatr Surg 2004;39:931–936. La Quaglia MP, et al. The impact of gross total resection on local control and survival in high-risk neuroblastoma. J Pediatr Surg 2004;39:412–417; discussion 412–417. Matthay KK, Perez C, Seeger RC, et al. Successful treatment of Stage III neuroblastoma based on prospective biologic staging: A Children’s Cancer Group study. J Clin Oncol 1998;16: 1256–1264. Gillis AM, Sutton E, Dewitt KD, et al. Long-term outcome and toxicities of intraoperative radiotherapy for high-risk neuroblastoma. Int J Radiat Oncol Biol Phys 2007;69:858–864. Children’s Oncology Group. Available at: http://www. childrensoncologygroup.org. Accessed June 2008. von Allmen D, Grupp S, Diller L, et al. Aggressive surgical therapy and radiotherapy for patients with high-risk neuroblastoma treated with rapid sequence tandem transplant. J Pediatr Surg 2005;40:936–941; discussion 941. von Schweinitz D, Hero B, Berthold F. The impact of surgical radicality on outcome in childhood neuroblastoma. Eur J Pediatr Surg 2002;12:402–409. Philip T, Ladenstein R, Zucker JM, et al. Double megatherapy and autologous bone marrow transplantation for advanced neuroblastoma: The LMCE2 study. Br J Cancer 1993;67:119–127. Grupp SA, Stern JW, Bunin N, et al. Tandem high-dose therapy in rapid sequence for children with high-risk neuroblastoma. J Clin Oncol 2000;18:2567–2575. Kletzel M, Katzenstein HM, Haut PR, et al. Treatment of highrisk neuroblastoma with triple-tandem high-dose therapy and stem-cell rescue: Results of the Chicago Pilot II Study. J Clin Oncol 2002;20:2284–2292.
doi:10.1016/j.ijrobp.2008.10.069
CLINICAL INVESTIGATION
Pediatrics
EXCELLENT LOCAL CONTROL FROM RADIATION THERAPY FOR HIGH-RISK NEUROBLASTOMA HEATHER G. GATCOMBE, M.D.,* R. B. MARCUS, JR., M.D.,x HOWARD M. KATZENSTEIN, M.D.,y MOURAD TIGHIOUART, PH.D.,z AND NATIA ESIASHVILI, M.D.* * Department of Radiation Oncology, y Children’s Healthcare of Atlanta, z Biostatistics and Informatics at Winship Cancer Institute, Emory University School of Medicine, Atlanta, GA; and x Department of Radiation Oncology, University of Florida, Gainesville, FL Purpose: Local recurrence has been demonstrated in previous studies to be one of the obstacles to cure in neuroblastoma. Radiation therapy indications, optimal dose, and technique are still evolving. Here we report our experience of high-risk neuroblastoma patients who received local radiation therapy as part of their cancer management. Methods and Materials: We conducted a retrospective study of 34 high-risk neuroblastoma patients who received radiation therapy to local sites of disease from March 2001 until February 2007 at our institution as part of their multimodality therapy. Results: At a median follow-up of 33.6 months, 6 patients died of disease, 7 patients were alive with disease, and 21 patients were in clinical remission. Eleven patients relapsed, all distantly. Two patients failed locally in addition to distant sites. Both of these patients had persistent gross disease after induction chemotherapy and surgery. Our 3-year local control, event-free survival, overall survival were 94%, 66%, and 86%, respectively. Conclusion: Patients with high-risk neuroblastoma in our series achieved excellent local control. Doses of 21–24 Gy to the primary tumor site appear to be adequate for local control for patients in the setting of minimal residual disease after induction chemotherapy and surgery. Patients with significant residual disease may benefit from radiation dose escalation, and this should be evaluated in a prospective clinical trial. Ó 2009 Elsevier Inc. High risk, Neuroblastoma, Local control, Radiation.
remain high in larger cooperative group studies. The Children’s Oncology Group (COG) in its current high-risk protocol will evaluate dose escalation in patients with gross disease to a dose of 36 Gy, whereas patients with microscopic residual disease will continue to receive 21.6 Gy. It is important to identify the subset of patients with a higher risk of local failure and determine whether they would benefit from dose escalation. The question of dose response on local tumor control is especially crucial in this young patient population in which radiation late effects are of high concern. Herein, we report our institutional experience in treating high-risk neuroblastoma with multimodality therapy incorporating local external beam radiation therapy (EBRT) with specific focus on local disease control. In addition to known patient and tumor-related risk factors, we analyzed whether radiation therapy dose and extent of surgical resection affected local disease control.
INTRODUCTION Neuroblastoma is the most common extracranial solid tumor of childhood. Approximately 50% of newly diagnosed patients have high-risk disease, and despite aggressive multimodality treatment incorporating induction chemotherapy, maximum tumor resection, myeloablative treatments, and maintenance therapy with cis-retinoic acid, and monoclonal antibody, these children do poorly with long-term survival rates of less than 40% (1–6). Although distant relapse is the main obstacle to cure in these patients, local failure remains a significant problem. The addition of radiation therapy to the multimodality approach of these high-risk patients has resulted in improvement in local control as supported from a large cooperative group study and several institutional reports (7–9). The optimal radiation therapy indications, technique, and dose are still evolving. Although there have been no randomized trials of radiation therapy in patients with high-risk disease, there was evidence of a dose response in the Children’s Cancer Group (CCG-3891) trial (7). Local failure rates, however,
METHODS AND MATERIALS After receiving internal review board approval, we conducted a retrospective chart review of all high-risk neuroblastoma patients
Reprint requests to: Natia Esiashvili, M.D., Department of Radiation Oncology, Emory School of Medicine, 1365 Clifton Road, NE, Atlanta, GA 30322. Tel: (404) 778-5782; FAX: (404) 7784139; E-mail: [email protected]
Conflict of interest: none. Received Aug 5, 2008, and in revised form Oct 15, 2008. Accepted for publication Oct 16, 2008. 1549
1550
I. J. Radiation Oncology d Biology d Physics
who received radiation therapy in our department from 2001 to 2007. All 34 patients were International Neuroblastoma Staging System (INSS) Stage 3 and 4 high-risk neuroblastoma as defined by the COG Risk Stratification Schema. All patients had histologic confirmation of their disease with either biopsy of their primary or metastatic tumor site or the presence of neuroblastoma in the bone marrow with elevated urinary catecholamines. The majority of patients were enrolled in the COG neuroblastoma biology study to determine biologic variables. Staging workup was performed with imaging including computed tomography (CT) of the chest and abdomen, 123I-meta-iodobenzylguanidine (MIBG) scan, technetium-99 bone scan, and bilateral bone marrow aspirates and biopsy. Patients were staged according to INSS criteria and assigned to the high-risk group according to COG risk-stratification scheme (Stage 4 disease in patients older than 1 year, MYCN-amplified tumors in patients with Stage 2, 3, or 4S disease or with Stage 4 disease and younger than 1 year, or unfavorable histology in patients older than 1 year with Stage 3 disease) (10). The majority of patients were treated either on or according to recent COG protocols ANBLOOP1 and ANBL02P1, including induction chemotherapy with cyclophosphamide, doxorubicin, vincristine, cisplatin, etoposide, and topotecan. Patient response was evaluated by bone marrow, bone scan, MIBG, and CT. Surgery was attempted on all patients unless they had a complete or near-complete response to chemotherapy. The goal of surgery was a gross total resection with preservation of vital organs. Near or gross total resection (NTR/GTR) was defined as > 95% surgical resection. Patients then underwent either single or tandem autologous stem-cell transplantation following high-dose chemotherapy. After recovery from myelosuppression, patients were treated with EBRT to primary tumor sites regardless of the extent of surgical resection and metastatic sites as indicated on the basis of their chemotherapy response. CT planning was used to define the treatment volume with customized blocking. EBRT was delivered by linear accelerator, 6-MV energy. The gross tumor volume (GTV) consisted of the postchemotherapy, presurgical disease. For abdominal primaries, the clinical tumor volume (CTV) incorporated the para-aortic lymph nodes in addition to GTV. The planning target volume (PTV) was created by expanding the CTV 1–2 cm, taking into account anatomic constraints. In most cases, opposed anterior and posterior fields were used. An intensity-modulated radiation therapy (IMRT) plan was performed for six patients with similar guidelines for targeting primary tumor and regional lymph nodes in the PTV and appropriate dose constraints to protect normal tissues. Daily fraction size was 1.5, 1.8, or 2 Gy daily and, in one case, 1.2 Gy twice daily. Six patients were simultaneously treated to metastatic sites of disease for persistent metastatic disease as determined from postinduction MIBG and bone-scan positivity. The decision to irradiate sites of distant metastatic disease was at the discretion of the treating physician and parental choice. Three patients received total body irradiation (TBI) for transplant conditioning consisting of 1.5 Gy twice daily for 4 days with at least a 6-hour separation between morning and afternoon fractions. Lung blocks were used to decrease the lung dose. After completion of radiation therapy, patients received adjuvant cis-retinoic acid for six cycles as part of their maintenance therapy for treating minimal residual disease. Patients were followed after completion of treatment in the pediatric hematology/oncology and radiation oncology clinics with clinical examinations and appropriate laboratory and imaging testing to evaluate disease control. All of the disease relapses were documented with imaging and correlated with known adverse disease factors (MYCN amplification, Shimada histopathology), as well as with response to induction che-
Volume 74, Number 5, 2009
motherapy, absence or presence of residual disease in the primary site after surgery, number of autologous bone marrow transplants, and radiotherapy parameters. Diagnostic imaging documenting relapse was compared with RT fields and dosimetry to determine failure inside or outside of treatment fields.
Statistical analysis Relapse-free survival was calculated from the date of diagnosis to date of relapse or progression or to date of last follow-up if patients remained in remission. Local relapse-free survival was calculated from date of diagnosis to date of local failure at primary tumor site as seen on CT, MIBG, or bone scan. Overall survival was calculated from date of diagnosis to date of last follow-up or date of death. Kaplan-Meier statistics were used to calculate actuarial disease-free, overall, and local relapse-free survival curves.
RESULTS From March 2001 until February 2007, we treated 34 (15 male, 19 female) children with high-risk neuroblastoma with local radiation to their primary site of disease. Patient characteristics are summarized in Table 1. Median age at diagnosis was 3.4 years (range, 10 months–20 years). Five patients had INSS Stage 3 and 29 Stage 4 neuroblastoma. The primary tumor site was abdominal in 27 patients, thoracic in 3 patients, pelvic in 2 patients, and combined in 2 patients. MYCN was amplified in 12 patients, nonamplified in 16, and unknown in 6 patients. Shimada classification was unfavorable in 26, favorable in 3, and unknown in 5. Following induction chemotherapy, 24 patients had persistent disease, and 10 demonstrated a complete response as determined by postinduction scans. Surgery was performed after induction chemotherapy and before transplantation. Eighteen patients had a gross or near total resection (GTR/ NTR), 12 achieved a subtotal resection (STR), 1 patient had a biopsy alone, and 3 did not undergo surgery. The patient who had a biopsy alone had unresectable disease due to involvement of the aorta, inferior vena cava, and superior mesenteric artery. The 3 patients who did not undergo a surgical resection had a complete or near complete response to induction chemotherapy on the basis of radiographic and functional imaging. The remaining 7 patients who had a complete response following chemotherapy based on functional imaging (MIBG and bone scan) still had persistent small-volume residual disease on CT scan for which they were taken to surgery. Following resection, 6 patients underwent a single transplant and 28 underwent tandem transplant. The reasons for undergoing a single transplant only were complications from surgery or chemotherapy (2 patients), Stage 3 disease (1 patient), inadequate stem-cell counts for tandem transplant (1 patient), and parental refusal (1 patient); another patient failed to clear his bone marrow and went on to receive MIBG therapy followed by a single transplant. Three patients received total body irradiation (TBI) in preparation for transplantation. These patients were treated according to bone marrow transplant protocols adopted before March 2002 and received 12 Gy at 1.5 Gy twice daily. The remaining
RT for high-risk neuroblastoma d H. G. GATCOMBE et al.
Table 1. Patient Characteristics (34 patients) Age Median Range Sex (n) Male Female INSS stage 3 4 Primary site Abdomen Thorax Pelvis Combined Metastatic sites Skeletal Bone marrow Distant lymph node Orbit Lung Response to induction chemo (based on MIBG and bone scan) CR Persistent disease Resection extent GTR/NTR STR Biopsy only No surgery Single transplant Tandem transplant
Table 2. Radiation Dose Radiation dose (Gy)
3.4 years 10 months–20 years 15 19 5 29
1551
Primary site dose (excluding TBI ; n = 34)* 12–18 19–24 25–30.6 Metastatic sites (n = 6) 21–22 22–23 23–24
n (% patients) 3 (9) 27 (79) 4 (2) 4 (67) 1 (17) 1 (17)
27 3 2 2
Abbreviation: TBI = total body irradiation. * Three patients received TBI to 12 Gy (1.5 Gy twice daily) in addition to primary site radiation doses of 12, 18, and 18 Gy.
28 23 9 1 2
3-year overall survival was 86%. Our 3-year local control was 94% (Fig. 2).
PATTERNS OF RELAPSE 10 (29%) 24 (71%) 18 12 1 3 6 28
Abbreviations: CR = complete response; GTR = gross total resection; INSS = International Neuroblastoma Staging System; MIBG = 123I-meta-iodobenzylguanidine; NTR = near total resection; STR = subtotal resection.
31 patients received high-dose chemotherapy for myeloablative conditioning. The median time from last transplant until the start of radiation was 61.5 days (range, 38–136 days). All 34 patients received radiation to the primary disease site with the majority treated to the abdomen (27 abdominal, 3 thoracic, 2 pelvic, and 2 combined abdominal and thoracic fields). The median radiation dose was 22 Gy (range, 18.0– 30.6), which includes the cumulative dose for the 3 patients who additionally received TBI (Table 2). The most common fraction schemes used were 21.6 Gy at 1.8 Gy per day (10 patients) and 21 Gy at 1.5 Gy per day (8 patients). During radiotherapy to their primary site of disease, 6 patients received concurrent RT to persistent sites of metastatic sites of disease (dose range, 21 Gy at 1.5 Gy/fraction to 30.6 Gy at 1.8 Gy/ fraction) as seen on postinduction scans (Table 3). The majority of patients required general anesthesia because of their young age. All but 1 patient received adjuvant cis-retinoic acid as part of their maintenance therapy. The median follow-up was 33.6 months (range, 0.5–70.5 months). At the time of analysis, 21 patients had no evidence of disease (NED), 7 were alive with disease (AWD), and 6 were dead of disease (DOD). Eleven of the 34 patients relapsed, all distantly and 2 with concurrent local failure. The 3-year event-free survival (EFS) was 66% (Fig. 1), and
Sites of distant relapse included skeletal (8 patients), bone marrow (5 patients), lung (4 patients), liver (1 patient), leptomeningeal (1 patient), and face/orbit (1 patient). The majority of patients who relapsed in bone and bone marrow had disease present there at initial presentation. The majority of patients with lung, liver, and leptomeningeal relapse did not present with disease at these locations at initial diagnosis. Of the 11 patients who relapsed, all but one were INSS Stage 4. MYCN amplification and histology type in our series was not strongly associated with the risk of relapse. Seven of the 11 relapsed patients demonstrated distant persistent disease on MIBG or bone scan following induction chemotherapy. One of the 11 did not have a bone or MIBG scan postinduction, but CT demonstrated a residual primary tumor mass measuring 6 8cm. Four of the 11 patients had undergone a GTR/NTR, 6 had a STR, and 1 had a biopsy only. Extent of resection and response to induction chemotherapy were found to have no statistically significant influence on EFS. The majority of relapsed patients did not receive radiotherapy to metastatic sites. Of 6 patients treated to metastatic lesions, only 1 suffered progression inside the radiation field. When examined by the number of transplantations completed, patients who received tandem stem-cell rescue seemed to have a better outcome than patients receiving only a single stem-cell rescue (p = 0.043, Fig. 3). The median age of patients undergoing a single or tandem stem-cell rescue was significantly different (6.3 and 2.9 years, respectively). Moreover, the younger patients were more likely to undergo tandem stem-cell rescue and overall had better outcome (HR 1.18, p = 0.0256). After age was taken into account, the number of stem cell rescues were no longer significant. Both patients with local failure were INSS Stage 4 with unfavorable histology and MYCN amplification. They both presented with bone marrow and skeletal metastases at diagnosis and had persistent disease seen after induction chemotherapy (one locally and the other distantly). Both underwent
1552
I. J. Radiation Oncology d Biology d Physics
Volume 74, Number 5, 2009
Table 3. Metastatic disease location and local control of irradiated metastatic lesions
Bone Distant lymph nodes Other
No. of Patients with metastatic disease at Presentation*
No. of Patients with metastatic disease after induction chemotherapy*
No. of metastatic sites irradiated
Local control of irradiated metastatic sites
28 9 3
13 0 1
3 2y 1
2 2 1
* By 123I-meta-iodobenzylguanidine and bone scan. y Both patients had persistent enlarged supraclavicular lymph nodes on CT imaging after induction chemotherapy.
1.00
Event Free Survival
0.75
3-year Event-Free Survival 66%
0.50
High-risk neuroblastoma patients continue to present many challenges for treatment. These children often present with widely metastatic disease and large, bulky, invasive tumors involving critical vasculature and a gross resection is often difficult to obtain. Despite aggressive multimodality therapy including intensification of systemic therapy, local relapse remains an obstacle to cure in high-risk neuroblastoma patients. Adjuvant radiation therapy with EBRT, intraoperative radiotherapy (IORT), or both has been incorporated into high-risk neuroblastoma protocols to improve local control, but the optimal dose to deliver to these patients is unknown (7, 9, 11–13). Careful consideration of radiation dose and volume is important given the young age of these patients and the abutting critical normal structures. What is the optimal dose for these young patients? The addition of local radiation therapy to aggressive systemic therapy and tumor resection has resulted in good local control rates in institutional studies. In our study group, the local control at the primary tumor site was 94% at 3 years. Our results are consistent with previous reports of high-risk patients treated with aggressive multimodality therapy including double or triple transplantation and incorporating local radiation
0.25
DISCUSSION
therapy (8, 9, 12, 14). Marcus et al. (9) reported an actuarial probability of local control of 97% with a median follow-up of 19.4 months in patients with Stage 4 or high-risk Stage 3 neuroblastoma treated with induction chemotherapy, tandem transplant, surgery, and local radiotherapy (12 Gy TBI plus local EBRT dose 10.5–18 Gy). Bradfield et al. (14) reported a local failure rate of 7% (95% confidence interval, 0%–14%) at 2 years after delivering 21 Gy at 1.5 Gy a day. Kushner et al. (8) reported primary site failure rates of 4.1% 3.0% at 12 months and 10.1% 5.3% at 36 months from the start of radiation; their protocol consisted of 21 Gy in 1.5-Gy fractions twice daily to all high-risk neuroblastoma patients, regardless of resection status. The Chicago Pilot II Protocol achieved a local control rate of 97% in a cohort of 30 highrisk patients with abdominal primaries who received 24 Gy and triple-tandem high-dose therapy (12). Taken together, the data may suggest that doses of 21–24 Gy are adequate for local control for the majority of patients with high-risk neuroblastoma, but these results were hard to reproduce in the cooperative group setting. In our study, only six patients were treated with IMRT, and the majority received conventional radiotherapy. Evaluation of various radiation delivery techniques have been previously reported from our institution in an effort to limit the radiation dose to normal tissues and decrease potential toxicities (15). IMRT was not found to be beneficial in patients with lateralized tumors because it delivered a higher radiation dose to the spleen, liver, and stomach and contralateral kidney and was used less frequently in more recently treated patients.
0.00
a STR with gross residual disease left at the primary site. The first patient underwent tandem stem-cell rescue, with radiation starting 62 days after his second rescue. He had 1.3 cm of residual disease as documented on CT before initiation of radiotherapy. The second patient had a biopsy only for a large tumor encasing the aorta, inferior vena cava, and superior mesenteric artery measuring 2.8 8.0 6.0 cm on CT. He received tandem stem-cell rescue with local radiation starting 108 days after his last rescue. The first patient received 22.5 Gy, and the second patient received 12Gy TBI and then 12 Gy local radiation to residual disease. Both patients suffered from both local-regional and distant failures simultaneously and died of disease. Of the 6 patients who also received concurrent radiation to metastatic sites of disease, there was only one treatment failure inside the radiotherapy field. A patient who received 21.6 Gy to a left femur metastasis recurred at this site in addition to lung and other skeletal sites 38 months after completing radiation. An additional patient continues to exhibit a persistent abnormality on bone scan after receiving 23.4 Gy to the calvarium, but this area at last follow-up has remained stable without evidence of disease progression or relapse.
0
20
40
Time (months)
Fig. 1. Event-free survival.
60
80
RT for high-risk neuroblastoma d H. G. GATCOMBE et al.
1.00
Local Control
1553
Relapse Free Survival Kaplan-Meier Curve
100
3-year Local Control 94%
Numb Transp 1 2
0.75
90
0.25
0.50
Percent
80 70 60 50
0.00
40 0 0
20
40
60
80
Time (months)
10
20
30
40
50
60
70
80
Time to Relapse (In months)
Fig. 2. Local control.
Fig. 3. Event-free survival by number of transplants.
The optimal radiation target volume is still disputable. In our series, consistent with other institutional protocols, we included the postinduction chemotherapy and presurgical tumor volume plus a 2-cm margin. However, we also elected to cover the first-echelon draining lymph nodes for all patients with abdominal primaries. Some groups argue against including the uninvolved draining lymph nodes within the radiation and field, and it was not allowed on the preceding COG protocols. It will be important to know whether this had any impact on disease control—specifically on the incidence of regional nodal failures, which are sometimes hard to distinguish anatomically from primary site recurrence. Of our 11 patients who relapsed, 4 received a GTR/NTR, 6 had an STR, and 1 patient had a biopsy alone. There have been conflicting reports of whether extent of resection improves outcome. Browne et al. (12) found that although patients with a GTR had lower overall recurrence rates, there was no significant difference in the local recurrence rate. Analysis of the CCG-3891 trial demonstrated no improvement in event-free or overall survival by achieving a complete resection (16). Alternatively, La Quaglia et al. (17) found a 50% probability of local progression in unresected patients compared with 10% in patients undergoing GTR (p <.01). Their overall survival rate in resected patients was 50% compared with 11% in unresected patients (p < .01), and they recommended that GTR should be part of the management of Stage 4 neuroblastoma in patients older than 1 year (17). Matthay et al. (18) demonstrated a survival improvement for Stage 3 tumors with unfavorable biology with complete or near complete resection. Gillis et al. (19) also found a significant difference in patients undergoing GTR or STR who had been treated with IORT (19). Although the extent of surgical resection on EFS did not reach statistical significance, we emphasize the importance of achieving maximal surgical resection, given that, in our cohort, two patients who had significant residual disease at the primary site experienced local failure. These children went on to receive doses of 22.5 Gy (1.5 Gy daily) and 24 Gy (12 Gy TBI and 12 Gy to local site). Would patients with persistent gross disease benefit
from dose escalation? Further investigation is necessary to answer this question. An upcoming COG study plans on delivering doses of 21.6 Gy after GTR and 36 Gy to residual disease (20) . Aggressive initial surgical resection or second-look surgery is another consideration to improve local control, but, as with radiotherapy, it would require a careful balance between benefit and the surgical morbidity risk, which remains high (21, 22). One of the obstacles in evaluating the extent of surgery has been assessing the completeness of surgical resection. Von Allmen et al. (21) has reported a 48% discrepancy between the surgeons’ assessment of residual tumor and the imaging findings. Reliable methods for detecting viable residual tumor vs. mature scar tissue are essential for accurate selection of appropriate candidates for radiotherapy dose escalation. Our data show that even after STR, local control was possible in the majority of patients. The amount of residual tumor measured volumetrically may also have some significance and warrants careful evaluation in a cooperative group setting. Radiation therapy indications for neuroblastoma metastatic sites are also not standard. Most institutional reports, in fact, do not provide sufficient information on treating metastatic sites. The cooperative group studies including those recently completed and currently open are recommending radiation to persistent sites of metastatic disease as seen on MIBG and bone scans. In our patient population, most patients were treated according to COG criteria, and only one of our patients failed at an irradiated metastatic site (an additional patient continues to have a persistent abnormality on bone scan). The majority of patients who relapsed with skeletal metastasis had disease present at these locations at the time of initial diagnosis and could possibly have benefited from radiation to these sites regardless of responsiveness to induction chemotherapy. This question still warrants careful evaluation in prospective fashion. It is well-established that intensive chemotherapy including myeloablative high-dose chemotherapy with autologous stem-cell rescue adds to disease control. The number of rescues may have potential influence on disease outcome
1554
I. J. Radiation Oncology d Biology d Physics
(23–25). In our analysis, patients who underwent tandem stem cell rescue had a better outcome than those who received a single rescue. However, after this variable was adjusted for age, the number of transplants were no longer significant. Only two patients did not receive a second rescue because of complications from chemotherapy or surgery, and it is thus difficult to ascertain whether younger patients were able to tolerate the second rescue better or they have a better prognosis regardless of therapy type. In conclusion, radiation therapy continues to play an important role in the local control of high-risk neuroblastoma.
Volume 74, Number 5, 2009
Our institution has achieved excellent local control in our high-risk neuroblastoma patients. Doses of 21–24 Gy to the primary tumor site appear to be adequate for local control for patients in the setting of minimal residual disease. Although in our series most patients with residual disease still achieved good local control, patients with significant residual tumor may benefit from radiation dose escalation. A prospective randomized trial is necessary to determine whether patients with persistent gross disease after induction chemotherapy and surgery would benefit from further dose escalation.
REFERENCES 1. De Bernardi B, Nicolas B, Boni L, et al. Disseminated neuroblastoma in children older than one year at diagnosis: Comparable results with three consecutive high-dose protocols adopted by the Italian Co-Operative Group for Neuroblastoma. J Clin Oncol 2003;21:1592–1601. 2. Matthay KK, Villablanca JG, Seeger RC, et al. Treatment of high-risk neuroblastoma with intensive chemotherapy, radiotherapy, autologous bone marrow transplantation, and 13-cisretinoic acid. Children’s Cancer Group. N Engl J Med 1999; 341:1165–1173. 3. Matthay KK. Impact of myeloablative therapy with bone marrow transplantation in advanced neuroblastoma. Bone Marrow Transplant 1996;18(Suppl. 3):S21–S24. 4. Kushner BH, LaQuaglia MP, Bonilla MA, et al. Highly effective induction therapy for stage 4 neuroblastoma in children over 1 year of age. J Clin Oncol 1994;12:2607–2613. 5. Cheung NK, Kushner BH, Cheung IY, et al. Anti-G(D2) antibody treatment of minimal residual stage 4 neuroblastoma diagnosed at more than 1 year of age. J Clin Oncol 1998;16: 3053–3060. 6. Stram DO, Matthay KK, O’Leary M, et al. Consolidation chemoradiotherapy and autologous bone marrow transplantation versus continued chemotherapy for metastatic neuroblastoma: a report of two concurrent Children’s Cancer Group studies. J Clin Oncol 1996;14:2417–2426. 7. Haas-Kogan DA, Swift PS, Selch M, et al. Impact of radiotherapy for high-risk neuroblastoma: A Children’s Cancer Group study. Int J Radiat Oncol Biol Phys 2003;56:28–39. 8. Kushner BH, Wolden S, LaQuaglia MP, et al. Hyperfractionated low-dose radiotherapy for high-risk neuroblastoma after intensive chemotherapy and surgery. J Clin Oncol 2001;19: 2821–2828. 9. Marcus KJ, Shamberger R, Litman H, et al. Primary tumor control in patients with Stage 3/4 unfavorable neuroblastoma treated with tandem double autologous stem cell transplants. J Pediatr Hematol Oncol 2003;25:934–940. 10. Brodeur GM, Kletzel M, Cohn SL, et al. Revisions of the international criteria for neuroblastoma diagnosis, staging, and response to treatment. J Clin Oncol 1993;11:1466–1477. 11. Haas-Kogan DA, Fisch BM, Wara WM, et al. Intraoperative radiation therapy for high-risk pediatric neuroblastoma. Int J Radiat Oncol Biol Phys 2000;47:985–992. 12. Browne M, Kletzel M, Cohn SL, et al. Excellent local tumor control regardless of extent of surgical resection after treatment
13. 14.
15. 16. 17. 18.
19. 20. 21.
22. 23. 24. 25.
on the Chicago Pilot II protocol for neuroblastoma. J Pediatr Surg 2006;41:271–276. Zachariou Z, Sieverts H, Eble MJ, et al. IORT (intraoperative radiotherapy) in neuroblastoma: Experience and first results. Eur J Pediatr Surg 2002;12:251–254. Bradfield SM, Douglas JG, Hawkins DS, et al. Fractionated low-dose radiotherapy after myeloablative stem cell transplantation for local control in patients with high-risk neuroblastoma. Cancer 2004;100:1268–1275. Paulino AC, Ferenci MS, Chiang KY, et al. Comparison of conventional to intensity modulated radiation therapy for abdominal neuroblastoma. Pediatr Blood Cancer 2006;46:739–744. Adkins ES, Sawin R, Gerbing RB, et al. Efficacy of complete resection for high-risk neuroblastoma: A Children’s Cancer Group study. J Pediatr Surg 2004;39:931–936. La Quaglia MP, et al. The impact of gross total resection on local control and survival in high-risk neuroblastoma. J Pediatr Surg 2004;39:412–417; discussion 412–417. Matthay KK, Perez C, Seeger RC, et al. Successful treatment of Stage III neuroblastoma based on prospective biologic staging: A Children’s Cancer Group study. J Clin Oncol 1998;16: 1256–1264. Gillis AM, Sutton E, Dewitt KD, et al. Long-term outcome and toxicities of intraoperative radiotherapy for high-risk neuroblastoma. Int J Radiat Oncol Biol Phys 2007;69:858–864. Children’s Oncology Group. Available at: http://www. childrensoncologygroup.org. Accessed June 2008. von Allmen D, Grupp S, Diller L, et al. Aggressive surgical therapy and radiotherapy for patients with high-risk neuroblastoma treated with rapid sequence tandem transplant. J Pediatr Surg 2005;40:936–941; discussion 941. von Schweinitz D, Hero B, Berthold F. The impact of surgical radicality on outcome in childhood neuroblastoma. Eur J Pediatr Surg 2002;12:402–409. Philip T, Ladenstein R, Zucker JM, et al. Double megatherapy and autologous bone marrow transplantation for advanced neuroblastoma: The LMCE2 study. Br J Cancer 1993;67:119–127. Grupp SA, Stern JW, Bunin N, et al. Tandem high-dose therapy in rapid sequence for children with high-risk neuroblastoma. J Clin Oncol 2000;18:2567–2575. Kletzel M, Katzenstein HM, Haut PR, et al. Treatment of highrisk neuroblastoma with triple-tandem high-dose therapy and stem-cell rescue: Results of the Chicago Pilot II Study. J Clin Oncol 2002;20:2284–2292.