Advertisement
Journal of Clinical Oncology  
Search for:
Limit by:
  Browse by Subject or Issue
Home Search or Browse JCO My JCO Subscriptions Customer Service Site Map

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a colleague
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Save to my personal folders
Right arrow Download to citation manager
Right arrowRights & Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Seeger, R. C.
Right arrow Articles by Matthay, K. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Seeger, R. C.
Right arrow Articles by Matthay, K. K.
Journal of Clinical Oncology, Vol 18, Issue 24 (December), 2000: 4067-4076
© 2000 American Society for Clinical Oncology

Quantitative Tumor Cell Content of Bone Marrow and Blood as a Predictor of Outcome in Stage IV Neuroblastoma: A Children’s Cancer Group Study

By Robert C. Seeger, C. Patrick Reynolds, Richard Gallego, Daniel O. Stram, Robert B. Gerbing, Katherine K. Matthay

From the Departments of Pediatrics and Pathology, University of Southern California School of Medicine and Childrens Hospital; and Department of Preventive Medicine, University of Southern California School of Medicine, Los Angeles; Children’s Cancer Group, Arcadia; and Department of Pediatrics, University of California School of Medicine, San Francisco, CA.

Address reprint requests to Robert C. Seeger, MD, Children’s Cancer Group, PO Box 60012, Arcadia, CA 91066-6012; email rseeger{at}chla.usc.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 REFERENCES
 
PURPOSE: This study investigated the prognostic value of quantifying tumor cells in bone marrow and blood by immunocytology in children with high-risk, metastatic neuroblastoma.

PATIENTS AND METHODS: Patients with stage IV neuroblastoma (N = 466) registered on Children’s Cancer Group study 3891 received five cycles of induction chemotherapy and were randomized either to myeloablative chemoradiotherapy with autologous purged bone marrow rescue or to nonmyeloablative chemotherapy. Subsequently, they were randomized to 13-cis-retinoic acid or no further treatment. Immunocytologic analyses of bone marrow and blood were performed at diagnosis, week 4, week 12, bone marrow collection, and end induction and were correlated with tumor biology, clinical variables, treatment regimen, and event-free survival (EFS).

RESULTS: Immunocytology identified neuroblastoma cells in bone marrow of 81% at diagnosis, 55% at 4 weeks, 27% at 12 weeks, 19% at bone marrow collection, and 14% at end induction. Tumor cells were detected in blood of 58% at diagnosis and 5% at collection. There was an adverse effect on EFS of increasing tumor cell concentration in bone marrow at diagnosis (P = .04), at 12 weeks (P = .006), at bone marrow collection (P < .001), and at end induction (P = .07). Positive blood immunocytology at diagnosis was associated with decreased EFS (P = .003). The prognostic impact of immunocytology was independent of morphologically detected bone marrow disease, MYCN status, and serum ferritin level in bivariate Cox analyses.

CONCLUSION: Immunocytologic quantification of neuroblastoma cells in bone marrow and blood at diagnosis and in bone marrow during induction chemotherapy provides prognostic information that can identify patients with very high-risk disease who should be considered for experimental therapy that might improve outcome.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 REFERENCES
 
BONE MARROW involvement by neuroblastoma is extremely common in children with metastatic disease, present by standard morphologic examination of aspirates and biopsies in 60% to 65% of children with stage IV disease at diagnosis.1 Detection of tumor by immunocytology, using a mixture of monoclonal antibodies reactive at the cell surface, has been shown to reliably detect tumor with a sensitivity that may vary from 1 to 10 in 105 nucleated bone marrow cells, depending on the method of detection.2-14 Detection of small numbers of contaminating tumor cells in bone marrow or blood may be a very important component of response evaluation and is critical for evaluation of hematopoietic stem cell products, because the use of myeloablative therapy followed by autologous bone marrow transplantation now has been shown in a randomized trial to be beneficial to event-free survival (EFS).15

The goal of this study was to determine whether quantitative sequential assessment of bone marrow and peripheral-blood tumor cells was prognostic for disease response at the end of induction therapy and for ultimate EFS. We report here the correlation of bone marrow and blood immunocytology with outcome for all assessable patients with metastatic neuroblastoma treated on Children’s Cancer Group (CCG) protocol CCG-3891.15 This protocol was a phase III study using a standard induction chemotherapy followed by randomization to myeloablative chemoradiotherapy with purged autologous bone marrow transplantation (ABMT) or to intensive nonmyeloablative consolidation chemotherapy, and a second randomization to no further treatment or 6 months of 13-cis-retinoic acid biotherapy.


   PATIENTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
RESULTS
 DISCUSSION
 APPENDIX
 REFERENCES
 
Patients
Eligible patients included newly diagnosed children (1 to 18 years of age) enrolled on the CCG-3891 protocol for high-risk neuroblastoma. Protocol eligibility included patients with the following clinical and biologic characteristics: age >= 1 year and Evans stage IV disease (n = 434); age less than 1 year with stage IV and tumor MYCN gene amplification (n = 19). Thirteen additional children who had localized disease treated with surgery alone at diagnosis but who then developed metastases and enrolled on CCG-3891 were also included. Other high-risk patients enrolled on the CCG-3891 study but who lacked metastases were excluded from the analyses reported in this article in order to examine only children with stage IV disease. Signed informed consent by parents or guardians was obtained, along with appropriate local institutional review board approval. Patient accrual began January 1991 and ended April 1996.

Serum ferritin was measured by radioimmunoassay: levels >= 143 ng/mL were designated as unfavorable and levels less than 143 ng/mL were designated as favorable.16 Biologic features of tumor, including histology and MYCN gene copy number, were determined centrally in the CCG Neuroblastoma Biology Resource Laboratory. Tumor MYCN gene amplification was measured by Southern blot analysis (1991 through 1993) as previously described17 by both the pattern and intensity of MYCN protein expression by immunohistochemistry and by semiquantitative polymerase chain reaction (1994 through 1996).18 Central pathology review of all tumors was performed, with classification as favorable or unfavorable according to the Shimada classification system.19

Treatment
All patients received initial therapy with cisplatin 60 mg/m2 on day 0, doxorubicin 30 mg/m2 on day 2, etoposide 100 mg/m2 on days 2 and 5, and cyclophosphamide 1,000 mg/m2 on days 3 and 4 for five cycles at 28-day intervals, plus surgery and radiotherapy for gross residual disease. For patients randomized to ABMT, pretransplantation therapy consisted of continuous-infusion carboplatin 1,000 mg/m2/96 hours on days -8 to -5 and etoposide 640 mg/m2/96 hours on days -8 to -5, bolus melphalan 140 mg/m2 on day -7 and 70 mg/m2 on day -6, and total-body irradiation (3.33 Gy/d on days -3 to -1), followed by purged ABMT (median 2 x 108 mononuclear cells/kg) on day 0, followed by granulocyte colony-stimulating factor. The continuation chemotherapy arm used three cycles of continuous-infusion cisplatin 160 mg/m2/96 hours, etoposide 500 mg/m2/96 hours, and doxorubicin 40 mg/m2/96 hours simultaneously with bolus ifosfamide 2,500 mg/m2/d on days 0 through 3 and mesna followed by granulocyte colony-stimulating factor. The first randomization occurred just before the third cycle of initial induction therapy, at protocol week 8 (median, 59 days) for patients without disease progression. At the end of continuation therapy or ABMT (protocol week 34 from diagnosis; median, 285 days), patients without disease progression underwent a sequential second randomization to receive either six cycles of 13-cis-retinoic acid 160 mg/m2/d orally in two divided doses for 14 consecutive days in a 28-day cycle or no further therapy.15 Disease evaluations were performed at diagnosis and at the end of induction, after ABMT or continuation chemotherapy, and after retinoid therapy. Responses were assessed according to the International Neuroblastoma Response Criteria.20

Bone Marrow and Blood Immunocytology, Bone Marrow Collection, and Purging
Heparinized samples of peripheral blood and bilateral pooled posterior iliac crest bone marrow aspirates were sent to the CCG Neuroblastoma Resource Laboratory for analysis at the designated time points. Bone marrow was sent at diagnosis, 4 weeks (after one cycle of chemotherapy), 12 weeks (after three cycles of chemotherapy), precollection, at bone marrow collection, and at the end of the five cycles of induction chemotherapy. Peripheral blood was sent at diagnosis, week 12, and at collection. Samples were analyzed as previously described using a mixture of five monoclonal antibodies (HSAN1.2, 390, 459, 126-4, and BW575) with a sensitivity of one tumor cell per 105 nucleated cells.2

Bone marrow was collected for ABMT and processed at the CCG Neuroblastoma Purging Center just before the fourth or fifth cycle of initial therapy if a diagnostic bone marrow aspirate obtained a week before collection had less than 1% tumor by morphologic and immunocytologic analysis. Bone marrow was purged using sedimentation, filtration, and two cycles of immunomagnetic separation.21-23 All infused marrows were tumor-free by bone marrow immunocytology.

Statistical Design and Analysis
Life-table methods were used to estimate EFS.24 An event was defined as the first occurrence of any disease progression, second malignancy, or death from any cause. EFS was calculated from various time points corresponding to the times of sample acquisition. Thus EFS from time of study entry was considered when investigating the prognostic power of blood or bone marrow immunocytology from samples taken at entry, whereas EFS from time of collection was considered when evaluating the prognostic power of immunocytology of samples taken at collection. The log-rank statistic was used to compare the EFS probabilities between subgroups of patients, with subgroups defined according to level of blood or bone marrow tumor content. When more than two levels were considered, tests for trend in hazard were constructed using the approach of Thomas et al.25 Relative-risk analysis was performed using the regression method of Cox.26 Multivariate Cox models were used to assess the independence of predictive information coming from blood and bone marrow immunocytology relative to other known prognostic factors, including serum ferritin, MYCN amplification, and Shimada histopathology classification.15


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
DISCUSSION
 APPENDIX
 REFERENCES
 
Patients and Bone Marrow Immunocytology Samples
A total of 466 children with stage IV or previously untreated low-stage disease that had metastasized were enrolled on CCG-3891. Of these patients, 422 had at least one bone marrow sample available for immunocytologic analysis. A total of 2,626 bone marrow immunocytology samples were obtained from these patients from diagnosis through the end of consolidation therapy. Only those obtained at diagnosis (n = 267), 4 weeks from diagnosis (n = 262), 12 weeks from diagnosis (after three cycles of induction chemotherapy; n = 244), 1 week before collection (n = 171), from the collected bone marrow (n = 200), and at the end of induction, approximately 20 to 24 weeks from diagnosis (n = 138), were included in the analyses here. Twenty-one patients had samples at six time-points, 62 patients had samples at five time-points, 92 patients had samples at four time-points, 80 patients had samples at three time-points, 83 patients had samples at two time points, and 72 patients had only a single sample available. The precollection sample was not examined separately for EFS and other associations, because it was obtained in such close proximity to the bone marrow collection. Samples for blood immunocytology were also obtained at diagnosis (n = 174), 4 weeks (n = 23), 12 weeks (n = 98), and at the time of bone marrow collection (n = 158).

The quantitation of tumor by immunocytology at the various time points used for analysis is shown for bone marrow and blood in Table 1 and Fig 1. Patient characteristics and the percentage of patients with positive samples by characteristic at the various time points for both blood and bone marrow immunocytology are listed in Table 2. The median age for these patients was 2.9 years, with a typical profile of clinical and biologic risk factors for children with stage IV neuroblastoma >= 1 year of age. Thirty-nine percent of patients had MYCN-amplified tumors, 60% had elevated serum ferritin, 95% had unfavorable histology, 68% had bone metastases, and 80% had bone marrow metastases (by morphologic examination of aspirates and biopsies).


View this table:
[in this window]
[in a new window]
  		Table 1. Bone Marrow and Blood Immunocytology*
 


View larger version (18K):
[in this window]
[in a new window]
  		Fig 1. Quantification of neuroblastoma cells in bone marrow and peripheral blood by immunocytology at diagnosis and at bone marrow collection. Bone marrow at (A) diagnosis and (B) at collection; peripheral blood at (C) diagnosis and at (D) collection.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Patient Characteristics and Correlation With Immunocytologic Detection of Tumor Cells in Blood and Bone Marrow
 
Eighty-one percent of children had a positive bone marrow immunocytology at diagnosis, with a median bone marrow tumor burden for positive samples of 8,674 cells per 105 nucleated cells (Table 1). The concordance of light microscopic evidence of bone marrow metastases by the local institutional pathologist (which included bilateral aspirate and biopsy) with bone marrow immunocytology was 86%. Thirty-three bone marrow samples were negative by both evaluations, 196 were positive by both evaluations, and 38 (14%) were discordant. Eighteen marrow samples were positive by light microscopic evaluation and negative by immunocytology, whereas 20 were negative by light microscopic evaluation and positive by immunocytology. When the precollection and collection samples were compared using a quantitative cutoff of fewer than 100 versus >= 100 tumor cells per 105 bone marrow cells, they were 96% concordant in 171 patients. However, there was lower concordance (71%) when considering whether samples had only one or more tumor cell per 105 bone marrow cells, probably because of the much larger volume of the sample (100-fold) obtained at actual collection compared with that obtained at precollection.

The percentage of patients with any detectable bone marrow tumor by immunocytology steadily decreased during therapy, as shown in Fig 1, Table 1, and Table 2, from 81% of patients at diagnosis to 14% by the end of induction. If the cutoff of >= 100 tumor cells per 105 nucleated cells is used for the time points of diagnosis, 4 weeks, 12 weeks, collection, and end of induction therapy, the percentages positive were 69%, 23%, 6%, 4%, and 2%, respectively. In contrast to the results in bone marrow, no further decrement in circulating tumor was seen after 4 weeks (Table 1). Blood samples for immunocytology at diagnosis, week 4, week 12, and collection showed circulating tumor cells (>= one tumor cell per 105) at these times in 58%, 4%, 5%, and 5% of patients, respectively.

Children with high-risk stage III neuroblastoma were also entered on CCG-3891 (n = 72). Their results are not included in the analyses that follow in this article in order to have a uniform group of stage IV patients, but the immunocytology quantitation is given here for comparison. Immunocytology at diagnosis was positive in nine of 36 children with stage III disease, with a median tumor cell number of 7 per 105 nucleated cells, and blood immunocytology was positive in only two of 19 patients. Bone marrow immunocytology at 4 weeks was positive in two of 18, at 12 weeks in one of 27, at collection in 0 of 18, and at end induction in 0 of 23 patients. There was no significant difference in EFS by bone marrow immunocytology at diagnosis. Only one patient with stage III neuroblastoma had positive blood immunocytology at both 12 weeks and harvest, but not at the end of induction. This patient experienced relapse and died.

Immunocytology and Biologic and Clinical Risk Factors
Association of immunocytology with prognostic factors, sites of metastases and response to therapy were evaluated using the {chi}2 test of proportions (Table 2). Positive immunocytology in bone marrow at diagnosis was associated with both bone and morphologic bone marrow metastases, whereas positive immunocytology in blood at diagnosis was associated significantly only with morphologic bone marrow metastases. Positive blood immunocytology at diagnosis also was associated with the unfavorable biologic factors of elevated serum ferritin and tumor MYCN gene amplification. Negative bone marrow immunocytology at week 4 was associated with complete or very good partial response at the end of induction, whereas positive immunocytology at week 4 showed a trend (P = .065) toward association with partial response.

Immunocytology and Prognosis
The relationship of EFS to quantitative bone marrow immunocytology at time of diagnosis, 12 weeks, the time of collection, and end of induction just before consolidation therapy is shown in Fig 2. Results were similar for disease-free survival, with a slight decrease in significance (P = .069) for immunocytology at diagnosis. There was an adverse effect of increasing tumor cell concentration in bone marrow on EFS both at diagnosis (P = .038), 12 weeks (P = .006), and collection (P < .001). There was also a trend at the end of induction therapy (P = .072), whether the patient went on to ABMT or consolidation chemotherapy. The result of immunocytology at 4 weeks, after only one cycle of chemotherapy, was not associated with EFS (P = .468), even though it was correlated with induction response (P = .0178) (Table 2). A significant difference, listed in Table 3, was also seen in EFS at weeks 0 and 12, collection, and preconsolidation using a cutoff point of >= 100 tumor cells versus fewer than 100 tumor cells per 105 nucleated cells, with 3-year EFS at diagnosis for that level of tumor cells of 36% versus 22% (P = .018), 27% versus 7% at 12 weeks (P = .0002), 28% versus 0% at the time of collection (P < .0001), and 32% versus 0% at the end of induction therapy (P = .0014). Bone marrow immunocytology was also a highly significant (P < .0001) adverse factor using a cutoff tumor cell quantity of more than 20 tumor cells per 105 nucleated cells at the time of bone marrow collection, although only four patients had between 20 and 100 tumor cells at this time.



View larger version (26K):
[in this window]
[in a new window]
  		Fig 2. Neuroblastoma cells in bone marrow and EFS. Immunocytology was performed at (A) week 0 (diagnosis), (B) week 12, (C) the time of marrow collection, and (D) the end of induction chemotherapy. Patient groups with 0, 1 to 99, and >= 100 tumor cells per 105 nucleated bone marrow cells and the number of patients in each group are indicated.

 

View this table:
[in this window]
[in a new window]
  		Table 3. EFS According to Quantitative Bone Marrow Immunocytology and Treatment Arm
 
To test whether the patients with missing samples differed from those with immunocytology results, an analysis of EFS in these two subgroups was performed. There was no significant difference in EFS for patients with metastatic disease lacking immunocytology and those with this result available. Three-year EFS for those lacking an immunocytology sample at diagnosis compared with those with an immunocytology result using the log-rank test was 26% versus 27% (P = .21), 27% versus 25% at week 12 (P = .65), and 28% versus 32% at the end of induction (P = .19).

This adverse prognostic impact of bone marrow immunocytology >= 100 tumor cells per 105 nucleated cells at weeks 0, 12, or at collection remained significant (P < .05) in bivariate Cox analyses that included MYCN amplification or serum ferritin. Including Shimada classification as a predictor reduced the significance of the week 0 immunocytology in the analysis (P < .3), but this may be partly explained by reduced patient numbers, because 84 patients with immunocytology at diagnosis were missing the Shimada classification. After stratification by morphologic evidence of neuroblastoma in bone marrow, patients with >= 100 tumor cells in marrow at diagnosis had a significantly lower 3-year EFS (22%) than those with fewer than 100 tumor cells (44%; P = .016). However, for those patients with morphologically negative bone marrows at diagnosis (n = 53), the detection of >= 100 tumor cells by immunocytology was not predictive (P = .65). Nor was a difference seen (P = .8) when the patients who were negative by both measures were compared with those who were positive by light microscopy but negative by immunocytology (n = 20). At the end of induction, there were only five patients who were negative by light microscopy but positive on immunocytology, of whom three experienced relapse. Multivariate analysis of 139 patients with all four variables available, including immunocytology (>= 100 tumor cells at diagnosis), MYCN, ferritin, and Shimada classification, showed the same increased relative risk for immunocytology of 1.5 (95% confidence interval, 0.93 to 2.29) as in univariate analysis, but only MYCN retained a significant P value.

Positive blood immunocytology at week 0 but not at collection (Fig 3) was correlated with decreased EFS. In contrast to the bone marrow results, the blood immunocytology at week 0 was no longer predictive (P > .9) in a model that included MYCN amplification.



View larger version (20K):
[in this window]
[in a new window]
  		Fig 3. Neuroblastoma cells in peripheral blood and EFS. Immunocytology was performed at (A) diagnosis and (B) at the time of marrow collection. Patient groups with 0 or >= 1 (1+) tumor cells per 105 nucleated blood cells and the number of patients in each group are indicated.

 
When analyzed by treatment regimen, there was still an EFS advantage for patients with fewer than 100 tumor cells at diagnosis, 12 weeks, and at bone marrow collection, regardless of whether they were treated with ABMT or consolidation chemotherapy (Table 3). However, this was only significant (P < .05) in the 12-week (chemotherapy arm) or collection (ABMT arm) samples. There was not a significant difference in EFS for the subgroups of patients with either positive (>= one tumor cell per 105) or negative immunocytology at any of the time points by consolidation regimen. The only difference for immunocytology after stratification by postconsolidation therapy (± 13-cis-retinoic acid) was for patients who had positive immunocytology at collection and were randomized to no postconsolidation therapy. The four patients with positive immunocytology (>= one tumor cell) had a significantly lower EFS than those 52 with negative immunocytology (P = .037). After stratification by immunocytology, a relative though nonsignificant advantage in EFS was seen for those receiving 13-cis-retinoic acid, consistent with the overall benefit for this arm of the study.15

Next, in an attempt to discover whether the amount of change in immunocytology was significant, we examined EFS in patients whose paired bone marrow samples were grouped by changes from positive to negative compared with those who did not become negative. This analysis was limited by the numbers of patients who had paired samples at the various time points. For patients with any positive bone marrow immunocytology at diagnosis, regardless of the percentage, the EFS was not significantly different whether the tumor cells had cleared at 4 weeks or at 12 weeks. There was also no significant difference in the EFS for those patients whose immunocytology at 4 weeks was positive but at 12 weeks changed to negative, regardless of whether the cutoff of one tumor cell or 100 tumor cells was considered positive. We also examined the degree of decrease in marrow tumor cells by looking at groups who cleared 1, 2, or >= 3 logs of tumor cells between weeks 0 and 4 or weeks 0 and 12. No significant difference was seen on the basis of the degree of decrease in bone marrow tumor cells. This is in contrast to the fact that positive immunocytology at 12 weeks was a highly significant adverse prognostic factor when analyzed alone and may reflect the limited number of paired samples submitted for immunocytology.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
APPENDIX
 REFERENCES
 
Our data show that quantitative bone marrow immunocytology at diagnosis is a powerful prognostic factor in children with stage IV neuroblastoma. Moreover, this factor remains influential even after consideration of other known predictors of outcome, namely MYCN gene copy number,17 serum ferritin,16 and morphologic evidence of bone marrow involvement.1 This confirms and quantifies our previous observation that bone marrow is one of the unfavorable sites of metastases in stage IV neuroblastoma.1 We further have shown that patients with more than 100 tumor cells per 105 nucleated bone marrow cells after three to four cycles of chemotherapy had virtually no chance of survival, even with the myeloablative chemoradiotherapy and purged autologous bone marrow transplantation used in this study. Patients with a small number of persisting bone marrow tumor cells at the time of collection (< 100 per 105 nucleated bone marrow cells) did not have significantly different EFS than those with no detectable tumor. This suggests that the intensive pretransplantation conditioning or the consolidation therapy may be sufficient to eliminate small numbers of tumor cells, but not larger amounts. The fact that one half of the patients also received 13-cis-retinoic acid after consolidation may also have helped to eradicate small amounts of residual bone marrow tumor cells.15,27 However, the very small number of patients with detectable tumor cells in the bone marrow after consolidation or ABMT precluded the likelihood of seeing an impact of retinoid therapy on such patients.

It is also possible that the reason patients with small numbers of bone marrow tumor cells did not have a different EFS than those without immunocytologically detectable tumor was due to the influence of sampling variability on tumor detection. This sampling dependence was suggested by the lower concordance of immediate precollection and collection immunocytology results in patients with small amounts of tumor cells. Thus there may have been some patients in the so-called negative group who had small pockets of microscopic tumor.

Although the lack of immunocytology specimens for some patients at any given time point may have lessened the power of the analyses, the overall percentage of patients with metastatic disease on CCG-3891 who had a positive bone marrow at diagnosis by immunocytology (216 of 267 or 81%) or by morphology (374 of 466 or 80%) was similar, suggesting that there was nothing different about the group of patients for whom immunocytology results were not obtained. Furthermore, an analysis of EFS in patients with metastatic disease lacking immunocytology and those with this result available showed no significant difference at diagnosis (P = .21), week 12 (P = .65), or the end of induction (P = .19).

The rapidity of the bone marrow response did not seem to be prognostic, as evidenced by a failure to correlate EFS with bone marrow immunocytology at week 4 after a single cycle of combination chemotherapy. Neither the conversion from positive to negative nor the extent of the change seemed to influence prognosis. This is different from the results reported for childhood acute lymphoblastic leukemia, where a day 14 and even a day 7 induction phase bone marrow are prognostic.28

Our data clearly demonstrate in a very large population of patients with metastatic neuroblastoma that circulating tumor cells in peripheral blood at diagnosis are extremely frequent, occurring in 58% of children. Eight patients had more than 1% tumor cells among their total nucleated cells circulating in blood. This study supports the preliminary observations of others demonstrating circulating tumor cells in the blood or in peripheral-blood stem-cell collections of some patients with neuroblastoma, using immunologic detection, clonogenic assays, and reverse-transcriptase polymerase chain reactions (RT-PCR).29-35 Our data show that detection of circulating tumor cells in peripheral blood at diagnosis is a significant unfavorable prognostic factor and, furthermore, is correlated with other high-risk features, including MYCN amplification and elevated serum ferritin.

The data also show, not surprisingly, that circulating tumor cells are more likely to be found in patients with bone marrow metastases. Such peripheral-blood tumor cells persisted in a small percentage of patients even after multiple courses of chemotherapy, with little decrease subsequent to the first cycle of chemotherapy. This suggests that peripheral-blood stem-cell collections could be performed early in the course of treatment without increased risk of tumor contamination. However, the persistence of tumor cells in some patients emphasizes the importance of testing peripheral-blood stem-cell collections for tumor by sensitive methods before reinfusion and raises the question of the need for specific tumor-cell purging of stem-cell products. RT-PCR may increase the sensitivity of testing and provide an additional way to monitor circulating tumor cells.29,33-35 Parallel testing of RT-PCR, which may be an overly sensitive method, will be necessary together with immunocytology to determine which method is more valid for prediction of outcome.

In conclusion, quantifying tumor cells in peripheral blood and bone marrow at diagnosis and during induction therapy is an important prognostic factor for patients with stage IV neuroblastoma. Moreover, bone marrow immunocytology seems to provide independent prognostic information for this group of patients, beyond other known predictors such as MYCN copy number and serum ferritin. The frequency of circulating tumor cells also suggests the possible need for tumor purging when peripheral-blood stem cells are used for hematopoietic support. Failure to reduce tumor cells in bone marrow to fewer than 100 tumor cells per 105 bone marrow cells within 12 weeks of beginning treatment is a very unfavorable prognostic sign using current therapeutic approaches and may provide a surrogate marker to detect patients who should be changed from standard therapy to an alternative, innovative treatment with greater potential to improve outcome.


   APPENDIX
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX
REFERENCES
 
Go


View this table:
[in this window]
[in a new window]
  		Table 4. Participating Principal Investigators: Children’s Cancer Group
 
Go


View this table:
[in this window]
[in a new window]
  		Table 5. APPENDIX (Cont’d)
 

   ACKNOWLEDGMENTS
 
Supported by grants to the Children’s Cancer Group (CA13539) and R.C.S. (CA22794, CA02649, CA60104) from the Division of Cancer Treatment, National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Bethesda, MD; from the Neil Bogart Memorial Laboratories of the T.J. Martell Foundation for Leukemia, Cancer, and AIDS Research, Los Angeles, CA, and New York, NY (C.P.R., R.C.S.); and from the Campini Foundation, Neuroblastoma Research Fund, and Conner Fund, San Francisco, CA (K.K.M.).

We thank Carolyn Billups and Debra Collins for outstanding technical assistance.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 REFERENCES
 
1. DuBois SG, Kalika Y, Lukens JN, et al: Metastatic sites in stage IV and IVS neuroblastoma correlate with age, tumor biology, and survival [see comments]. J Pediatr Hematol Oncol 21: 181-189, 1999[Medline]

2. Moss TJ, Reynolds CP, Sather HN, et al: Prognostic value of immunocytologic detection of bone marrow metastases in neuroblastoma. N Engl J Med 324: 219-26, 1991[Abstract]

3. Evans AE, Griffin GC, Tartaglione M, et al: A method of detecting neuroblastoma in human bone marrow by means of two monoclonal antibodies PI 153/3 and KE2. Hybridoma 4: 289-296, 1985[Medline]

4. Cheung NK, Saarinen UM, Neely JE, et al: Monoclonal antibodies to a glycolipid antigen on human neuroblastoma cells. Cancer Res 45: 2642-2649, 1985[Abstract/Free Full Text]

5. Frantz CN, Duerst RE, Ryan DH, et al: A monoclonal anti-neuroblastoma antibody that discriminates between human nonhematopoietic and hematopoietic cell types. Hybridoma 5: 297-306, 1986[Medline]

6. Berthold F, Schneider A, Schumacher R, et al: Detection of minimal disease in bone marrow of neuroblastoma patients by immunofluorescence. Pediatr Hematol Oncol 6: 73-83, 1989[Medline]

7. Reid MM, Hamilton PJ: Histology of neuroblastoma involving bone marrow: The problem of detecting residual tumour after initiation of chemotherapy. Br J Haematol 69: 487-490, 1988[Medline]

8. Reid MM, Wallis JP, McGuckin AG, et al: Routine histological compared with immunohistological examination of bone marrow trephine biopsy specimens in disseminated neuroblastoma. J Clin Pathol 44: 483-486, 1991[Abstract/Free Full Text]

9. Caillaud JM, Martinez-Madrigal F, Hartmann O, et al: Immunohistochemical demonstration of neurone specific enolase in bone marrow infiltrated by neuroblastoma. J Clin Pathol 44: 309-312, 1991[Abstract/Free Full Text]

10. Papac RJ: Bone marrow metastases: A review. Cancer 74: 2403-2413, 1994[Medline]

11. Nagai J, Kigasawa H, Tomioka K, et al: Immunocytochemical detection of bone marrow-invasive neuroblastoma cells. Eur J Haematol 53: 74-77, 1994[Medline]

12. Matthay KK, Reynolds CP, Stram DO, et al: Quantitative tumor cell content of bone marrow and blood as an early predictor of response in high risk neuroblastoma: A Children’s Cancer Group study. Proc Am Soc Clin Oncol 16: 512a, 1997 (abstr 1844)

13. Cheung NK, Heller G, Kushner BH, et al: Detection of metastatic neuroblastoma in bone marrow: When is routine marrow histology insensitive? J Clin Oncol 15: 2807-2817, 1997[Abstract]

14. Brodeur GM: Are bone marrow aspirates or biopsies the preferred method to detect metastatic neuroblastoma to bone marrow? J Pediatr Hematol Oncol 20: 297-298, 1998 (editorial)[Medline]

15. Matthay KK, Villablanca JG, Seeger RC, et al: Treatment of high-risk neuroblastoma with intensive chemotherapy, radiotherapy, autologous bone marrow transplantation, and 13-cis-retinoic acid. N Engl J Med 341: 1165-1173, 1999[Abstract/Free Full Text]

16. Hann HW, 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[Abstract/Free Full Text]

17. 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]

18. Seeger RC, Wada R, Brodeur GM, et al: Expression of N-myc by neuroblastomas with one or multiple copies of the oncogene. Prog Clin Biol Res 271: 41-49, 1988[Medline]

19. Shimada H, Chatten J, Newton WA Jr, 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

20. Brodeur GM, Pritchard J, Berthold F, et al: Revisions of the international criteria for neuroblastoma diagnosis, staging, and response to treatment [see comments]. J Clin Oncol 11: 1466-1477, 1993[Abstract/Free Full Text]

21. Seeger RC, Reynolds CP: Treatment of high-risk solid tumors of childhood with intensive therapy and autologous bone marrow transplantation. Pediatr Clin North Am 38: 393-424, 1991[Medline]

22. Reynolds CP, Seeger RC, Vo DD, et al: Model system for removing neuroblastoma cells from bone marrow using monoclonal antibodies and magnetic immunobeads. Cancer Res 46: 5882-5886, 1986[Abstract/Free Full Text]

23. Matthay KK, Seeger RC, Reynolds CP, et al: Allogeneic versus autologous purged bone marrow transplantation for neuroblastoma: A report from the Childrens Cancer Group. J Clin Oncol 12: 2382-2389, 1994[Abstract/Free Full Text]

24. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: 457-481, 1958

25. Thomas DG, Breslow NE, Gart JJ: Trend and homogeneity analyses of proportions and life table data. Comput Biomed Res 10: 373-381, 1977[Medline]

26. Cox DR: Regression models and life tables. J R Stat Soc 34: 187-220, 1972

27. Reynolds CP, Kane DJ, Einhorn PA, et al: Response of neuroblastoma to retinoic acid in vitro and in vivo. Prog Clin Biol Res 366: 203-211, 1991[Medline]

28. Gaynon PS, Desai AA, Bostrom BC, et al: Early response to therapy and outcome in childhood acute lymphoblastic leukemia: A review. Cancer 80: 1717-1726, 1997[Medline]

29. Mattano LA Jr, Moss TJ, Emerson SG: Sensitive detection of rare circulating neuroblastoma cells by the reverse transcriptase-polymerase chain reaction. Cancer Res 52: 4701-4705, 1992[Abstract/Free Full Text]

30. Moss TJ, Cairo M, Santana VM, et al: Clonogenicity of circulating neuroblastoma cells: Implications regarding peripheral blood stem cell transplantation. Blood 83: 3085-3089, 1994[Abstract/Free Full Text]

31. Sanders DG, Wiley FM, Moss TJ: Serial immunocytologic analysis of blood for tumor cells in two patients with neuroblastoma. Cancer 67: 1423-1427, 1991[Medline]

32. Komada Y, Zhang XL, Zhou YW, et al: Flow cytometric analysis of peripheral blood and bone marrow for tumor cells in patients with neuroblastoma. Cancer 82: 591-599, 1998[Medline]

33. Yanagisawa TY, Sasahara Y, Fujie H, et al: Detection of the PGP9.5 and tyrosine hydroxylase mRNAs for minimal residual neuroblastoma cells in bone marrow and peripheral blood. Tohoku J Exp Med 184: 229-240, 1998[Medline]

34. Burchill SA, Bradbury FM, Selby P, et al: Early clinical evaluation of neuroblastoma cell detection by reverse transcriptase-polymerase chain reaction (RT-PCR) for tyrosine hydroxylase mRNA. Eur J Cancer 31A: 553-556, 1995

35. Tchirkov A, Kanold J, Giollant M, et al: Molecular monitoring of tumor cell contamination in leukapheresis products from stage IV neuroblastoma patients before and after positive CD34 selection. Med Pediatr Oncol 30: 228-232, 1998[Medline]

Submitted April 7, 2000; accepted July 7, 2000.




This article has been cited by other articles:


Home page
 JCOHome page
K. K. Matthay, G. Yanik, J. Messina, A. Quach, J. Huberty, S.-C. Cheng, J. Veatch, R. Goldsby, P. Brophy, L. S. Kersun, et al.
Phase II Study on the Effect of Disease Sites, Age, and Prior Therapy on Response to Iodine-131-Metaiodobenzylguanidine Therapy in Refractory Neuroblastoma
J. Clin. Oncol., March 20, 2007; 25(9): 1054 - 1060.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
K. K. Matthay, J. C. Tan, J. G. Villablanca, G. A. Yanik, J. Veatch, B. Franc, E. Twomey, B. Horn, C. P. Reynolds, S. Groshen, et al.
Phase I Dose Escalation of Iodine-131-Metaiodobenzylguanidine With Myeloablative Chemotherapy and Autologous Stem-Cell Transplantation in Refractory Neuroblastoma: A New Approaches to Neuroblastoma Therapy Consortium Study
J. Clin. Oncol., January 20, 2006; 24(3): 500 - 506.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
M. V. Corrias, L. B. Faulkner, A. Pistorio, C. Rosanda, F. Callea, M. S. L. Piccolo, P. Scaruffi, C. Marchi, L. Lacitignola, M. Occhino, et al.
Detection of Neuroblastoma Cells in Bone Marrow and Peripheral Blood by Different Techniques: Accuracy and Relationship with Clinical Features of Patients
Clin. Cancer Res., December 1, 2004; 10(23): 7978 - 7985.
[Abstract] [Full Text] [PDF]

Home page
 JNMHome page
B. H. Kushner
Neuroblastoma: A Disease Requiring a Multitude of Imaging Studies
J. Nucl. Med., July 1, 2004; 45(7): 1172 - 1188.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
I. Y. Cheung, M. S. Lo Piccolo, B. H. Kushner, and N.-K. V. Cheung
Early Molecular Response of Marrow Disease to Biologic Therapy Is Highly Prognostic in Neuroblastoma
J. Clin. Oncol., October 15, 2003; 21(20): 3853 - 3858.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Pathol.Home page
G. Mehes, A. Luegmayr, R. Kornmuller, I. M. Ambros, R. Ladenstein, H. Gadner, and P. F. Ambros
Detection of Disseminated Tumor Cells in Neuroblastoma: 3 Log Improvement in Sensitivity by Automatic Immunofluorescence plus FISH (AIPF) Analysis Compared with Classical Bone Marrow Cytology
Am. J. Pathol., August 1, 2003; 163(2): 393 - 399.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
K. K. Matthay, V. Edeline, J. Lumbroso, M. L. Tanguy, B. Asselain, J. M. Zucker, D. Valteau-Couanet, O. Hartmann, and J. Michon
Correlation of Early Metastatic Response by123I-Metaiodobenzylguanidine Scintigraphy With Overall Response and Event-Free Survival in Stage IV Neuroblastoma
J. Clin. Oncol., July 1, 2003; 21(13): 2486 - 2491.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
I. Y. Cheung, M. S. Lo Piccolo, B. H. Kushner, K. Kramer, and N.-K. V. Cheung
Quantitation of GD2 Synthase mRNA by Real-Time Reverse Transcriptase Polymerase Chain Reaction: Clinical Utility in Evaluating Adjuvant Therapy in Neuroblastoma
J. Clin. Oncol., March 15, 2003; 21(6): 1087 - 1093.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
C. Trager, P. Kogner, M. Lindskog, F. Ponthan, A. Kullman, and B. Kagedal
Quantitative Analysis of Tyrosine Hydroxylase mRNA for Sensitive Detection of Neuroblastoma Cells in Blood and Bone Marrow
Clin. Chem., January 1, 2003; 49(1): 104 - 112.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
L. M. Wagner, S. M. Guichard, R. A. Burger, C. L. Morton, C. M. Straign, R. A. Ashmun, L. C. Harris, P. J. Houghton, P. M. Potter, and M. K. Danks
Efficacy and Toxicity of a Virus-directed Enzyme Prodrug Therapy Purging Method: Preclinical Assessment and Application to Bone Marrow Samples from Neuroblastoma Patients
Cancer Res., September 1, 2002; 62(17): 5001 - 5007.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
G. A. Yanik, J. E. Levine, K. K. Matthay, J. C. Sisson, B. L. Shulkin, B. Shapiro, D. Hubers, S. Spalding, T. Braun, J. L.M. Ferrara, et al.
Pilot Study of Iodine-131-Metaiodobenzylguanidine in Combination With Myeloablative Chemotherapy and Autologous Stem-Cell Support for the Treatment of Neuroblastoma
J. Clin. Oncol., April 15, 2002; 20(8): 2142 - 2149.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
H. Geminder, O. Sagi-Assif, L. Goldberg, T. Meshel, G. Rechavi, I. P. Witz, and A. Ben-Baruch
A Possible Role for CXCR4 and Its Ligand, the CXC Chemokine Stromal Cell-Derived Factor-1, in the Development of Bone Marrow Metastases in Neuroblastoma
J. Immunol., October 15, 2001; 167(8): 4747 - 4757.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Pathol.Home page
D. S. B. Hoon, C. T. Kuo, S. Wen, H. Wang, L. Metelitsa, C. P. Reynolds, and R. C. Seeger
Ganglioside GM2/GD2 Synthetase mRNA Is a Marker for Detection of Infrequent Neuroblastoma Cells in Bone Marrow
Am. J. Pathol., August 1, 2001; 159(2): 493 - 500.
[Abstract] [Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a colleague
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Save to my personal folders
Right arrow Download to citation manager
Right arrowRights & Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Seeger, R. C.
Right arrow Articles by Matthay, K. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Seeger, R. C.