Surgical management in pediatric neuroblastoma diagnosis and treatment: a 20-year, single-center experience

Brianna Spencer · Akshilkumar Patel · Robert Cilley ,,3 · Christa N. Grant ,,3

Abstract

Keywords International Neuroblastoma Risk Group · International Neuroblastoma Staging System · MYCN ·Neuroblastoma · Resection

lntroduction

Childhood and adolescent cancers are major public health concerns, with an incidence rate of 171.01 per million in the United States [ 1]. As revolutionary new cancer treatments are created and implemented widely, expenditures for pediatric cancer are expected to rise by 27%, placing a greater financial burden on the American healthcare system [ 2]. Neuroblastoma is the most common extracranial solid tumor in infants and children; incidence has increased over time, so it remains a complex surgical problem for pediatric surgeons [ 3, 4].

Neuroblastomas arise from primordial embryonic neural crest cells of the peripheral nervous system, including the sympathetic ganglia and adrenal medulla [ 5]. Clinically, this leads to heterogeneity in presentation, response to treatment,and outcomes, with some tumors regressing significantly without treatment and others involving widespread metastases refractory to aggressive, multipronged treatments [ 6].Although the specific underlying etiology of neuroblastoma is unclear, somatic mutations account for more than 98% of cases [ 5]. Specifically,

MYCN

and

ALK

oncogene amplifications are associated with high-risk cases with poor prognoses; amplification of the

MYCN

oncogene is currently the most important biomarker [ 7].Clinical presentation of neuroblastoma varies by primary tumor site and can be classified into abdominal (adrenal and retroperitoneal) and extra-abdominal (thoracic, cervical, pelvic, and other) sites. Across multiple studies, cases presenting with extra-adrenal tumors have lower malignancy, better prognosis, less incidence of

MYCN

amplification, and more favorable clinical and biological characteristics compared to tumors arising in the adrenal gland [ 8– 10].The guidelines governing clinical approach to neuroblastoma treatment differ based on the patient’s disease stage. The International Neuroblastoma Staging System(INSS) was first developed in the 1980s and is a surgical staging system based on localization of the tumor, associated metastases, and surgical resectability at diagnosis [ 11].The Children’s Oncology Group developed a more robust risk stratification combining a patient’s (1) disease stage,(2) age at diagnosis, (3) tumor histopathology, (4)

MYCN

status, (5) DNA index, and (6) tumor ploidy, classifying patients into four risks groups from low to ultra-high risk[ 11– 13]. This International Neuroblastoma Risk Group(INRG) categorization, established in 2009, guides the current multimodality treatment approach to neuroblastoma.Surgery for neuroblastomas with image-defined risk factors(IDRFs) have been more successful and safer if preceded by chemotherapy to shrink tumor size [ 14]. Surgical morbidity and mortality increase with increasing IDRFs, so neoadjuvant chemotherapy is of benefit in most L2 and M tumors.The ideal number of cycles prior to definitive resection is not absolute, and response to therapy is variable and depends on multiple factors. A single center retrospective study stratified high risk neuroblastoma patients by the number of neoadjuvant chemotherapy cycles prior to surgery and found a significantly lower mortality in patients undergoing 4 cycles preoperatively compared to those who had only two cycles[ 15]. In a retrospective review detailing patients with IDRF,prolonged chemotherapy in these patients did not improve surgical resection [ 16]. In INRG high-risk neuroblastoma cases (INSS stage 1–3 with age > 18 months and

MYCN

amplification), patients who underwent extended surgical treatment with complete resection experienced higher event-free and overall survival rates, suggesting aggressive surgical approaches should be attempted in these patients[ 17]. Complete resection may be achieved by open or minimally invasive surgical (MIS) approaches. There has been increased utilization of MIS techniques in neuroblastoma for both biopsy and resection over time. While increased numbers of IDRFs are associated with increased surgical morbidity regardless of approach, MIS morbidity is low in well selected patient populations [ 18, 19].

We detail the results of our retrospective analysis of neuroblastoma patients diagnosed and treated at our rural tertiary pediatric hospital in the past 20 years with particular interest in biopsy technique and the role of surgery in neuroblastoma over time. The authors were interested in trends of the pediatric surgeon’s involvement in the care of neuroblastoma patients before and after the institution of the INRG staging system in 2009 and believed that the role of the surgeon would change when surgical resection was no longer a requirement for staging.

Methods

Following Institutional Review Board (IRB) approval(IRB#00014901), the institutional electronic medical record was queried using International Classification of Diseases-9 and -10 codes for neuroblastoma to generate a list of medical record numbers. The CoPath pathology information system(Cerner, Kansas City, MO, USA) was also queried to collect pathology reports on available tissue specimens from neuroblastoma patients. The data from both sources were used to create a database of neuroblastoma patients diagnosed and treated at our institution over the past 20 years, 1999–2019.The early decade spanned 1999–2008 and the later decade 2009–2019. Data collected include patient demographics,cancer treatment modality, survival, biopsy technique, surgical intervention, pathology staging, and

MYCN

gene amplification status where available. After all data points of interest were collected, trends in the data were analyzed using post hoc statistical analysis using analysis of variance (ANOVA)and Student’s

t

test.

Results

Patient demographics were similar in the early and late cohorts, except for fewer Black or African American patients treated in the more recent decade (13.0%, 3.4%,

P

= 0.03) as well as a slightly older average age at diagnosis (2.38, 3.65,

P

= 0.03) (Table 1). In the recent decade, there were less stage 3 neuroblastomas (15.2%, 5.2%,

P

= 0.04). There was no difference in treatment modalities, overall survival, recurrence, or mortality between the two decades. In terms of surgical involvement, there were fewer open adrenal neuroblastoma biopsies in the recent decade (45%, 26%,

P

= 0.04). For extra-adrenal tumors, there was no difference in the method of biopsy between the early and late groups, but there was a decrease in the percent of open surgeries in the late group,(77%, 31%,

P

= 0.01) with an increase in minimally invasive surgical resections (0, 31%,

P

= 0.02) (Table 2). While stage 1 tumors were resected within a month in both time periods,there were a greater number of days to surgery in the later cohort for the patients in INSS stage 2 and 2b (

P

= 0.03 and

P

< 0.01, respectively) (Table 3), and a shorter time to surgery for INSS stage 3 (did not reach significance) and INSS stage 4 tumors (

P

= 0.02). When all stages were combined,there was a shorter time to definitive surgery overall (117 vs. 91 days,

P

= 0.047).

Table 1 Patient demographics

Bolded values indicate statistical significance. International Neuroblastoma Staging System

Characteristics 1999–2009 cohort ( n = 46) 2010–2019 cohort ( n = 58) P (Student’s t test)Average age at diagnosis (y) 2.38 3.65 0.03 Sex, n (%)Male 31 (67.4) 32 (55.2) 0.10 Female 15 (32.6) 26 (44.8) 0.10 Race, n (%)White 36 (78.3) 44 (75.9) 0.38 Black or African American 6 (13.0) 2 (3.4) 0.03 Other race 2 (4.3) 9 (15.5) 0.03 Two or more races 1 (2.2) 3 (5.2) 0.22 Hispanic 1 (2.2) 0 (0) 0.12 Treatment modality, n (%)Surgery 40 (87.0) 49 (84.5) 0.35 Chemotherapy 35 (76.1) 47 (81.0) 0.26 Radiation 20 (43.5) 32 (55.2) 0.12 Surgery + chemotherapy + radiation 20 (43.5) 32 (55.2) 0.12 Survival, n (%)Alive 36 (78.3) 48 (82.8) 0.28 Deceased 10 (21.7) 10 (17.2) 0.28 Average survival (y) 7.13 8.03 INSS stage, n (%)Stage 1 8 (17.4) 7 (12.1) 0.22 Stage 2 4 (8.7) 6 (10.3) 0.39 Stage 2b 2 (4.3) 4 (6.9) 0.29 Stage 3 7 (15.2) 3 (5.2) 0.04 Stage 4 22 (47.8) 33 (56.9) 0.17 Stage 4s 1 (2.2) 2 (3.4) 0.35 No staging (congenital adrenal mass) 1 (2.2) 3 (5.2) 0.22 Positive metastases at diagnosis, n (%)Yes 28 (60.9) 43 (74.1) 0.08 Location of metastases Lymph nodes 11 (39.3) 11 (25.6) 0.07 Bone marrow 4 (14.3) 12 (27.9) 0.05 Liver 2 (7.1) 4 (9.3) 0.34 Skull 1 (3.6) 0 (0) 0.07 Multiple sites 8 (28.6) 16 (37.2) 0.11 Other sites 2 (7.1) 0 (0) 0.02 No 18 (39.1) 15 (25.9) 0.08 Recurrence after resection, n (%)Yes 11 (23.9) 18 (31.0) 0.21 No 26 (56.5) 31 (53.5) 0.38 Tumor not resected 6 (13.1) 9 (15.5) 0.37 Unknown 3 (6.5) 0 (0) 0.03 Primary tumor site, n (%)Skull 1 (2.2) 0 (0) 0.13 Mediastinum/thorax 10 (21.7) 14 (24.1) 0.39 Adrenal gland 22 (47.8) 33 (56.9) 0.18 Retroperitoneum 4 (8.7) 8 (13.8) 0.21 Abdominal mass 6 (13.1) 3 (5.2) 0.08 Pelvic 3 (6.5) 0 (0) 0.03 NMYC gene status, n (%)Amplified 5 (10.9) 11 (19.0) 0.13 Non-amplified 12 (26.1) 32 (55.2) < 0.01 Not available 29 (63.0) 15 (25.8) < 0.01

Table 2 Biopsy, surgical methods and outcomes

Bolded values indicate statistical significance

Modality 1999–2009 cohort ( n = 33) 2010–2019 cohort ( n = 42) P (Student’s t test)Biopsy (adrenal), n (%)Percutaneous 2 (6) 3 (7) 0.43 Open 15 (45) 11 (26) 0.04 Resection 8 (24) 15 (36) 0.13 Unknown 3 (6) 10 (17) 0.05 Bone marrow 5 (15) 3 (7) 0.13 Surgery (adrenal), n (%)Open 26 (79) 27 (64) 0.08 Minimally invasive 1 (3) 4 (9) 0.14 Unknown 1 (3) 6 (14) 0.05 1999–2009 cohort ( n = 13) 2010–2019 cohort ( n = 16) P (Student’s t test)Biopsy (extra adrenal), n (%)Percutaneous 2 (15) 6 (37) 0.098 Open 4 (31) 5 (31) 0.50 Resection 6 (46) 4 (25) 0.12 Unknown 1 (8) 1 (6) 0.40 Surgery (extra adrenal), n (%)Open 10 (77) 5 (31) 0.01 Minimally invasive 0 5 (31) 0.02 Unknown 0 2 (12) 0.10 1999–2009 cohort ( n = 33) 2010–2019 cohort (n = 42) P (Student’s t test)Outcomes (adrenal), n (%)Mortality 9 (27) 10 (24) 0.38 Recurrence 9 (27) 17 (40) 0.12 1999–2009 cohort ( n = 13) 2010–2019 cohort ( n = 16) P (Student’s t test)Outcomes (extra adrenal), n (%)Mortality 1 (8) 0 0.13 Recurrence 2 (15) 2 (12) 0.41

Table 3 International neuroblastoma staging system stage

Bolded values indicate statistical significance. International Neuroblastoma Staging System

INSS stage Average days to surgery (2000–2009) Average days to surgery (2010–2019) P (Student’s t test)1 ( n = 15) 19 15 0.30 2 ( n = 10) 3.5 94 0.03 2b ( n = 6) 5.5 52 < 0.01 3 ( n = 10) 100 58 0.16 4 ( n = 55) 163 109 0.02 4s ( n = 3) 237 12.5 < 0.01 Average 88 56.75 0.24 Time to definitive surgery (d) 117 91 0.047

Discussion

In this retrospective single institution study over a 20-year period, we found that there was overall no change in the role of the surgeon in the treatment of neuroblastoma after the adoption of the INRG staging system. We did find a decreased rate of open biopsies in the later decade for adrenal neuroblastomas. This finding could be due to improved interventional radiology staffing and infrastructure at our pediatric hospital within a larger healthcare system, and increased comfort level and accuracy with percutaneous biopsies in children. Our data is in line with recent literature supporting increased use of minimally invasive biopsy techniques which have similar diagnostic yield and at least equivalent safety profiles compared to traditional open biopsies [ 20]. It is also likely that with the shift away from upfront surgery for staging in the INSS era, attempts at initial complete resection decreased when IDRFs were present in the INRG staging era. This trend followed several studies demonstrating that patients with IDRFs have fewer complications and a higher likelihood of complete resection if surgery is preceded by chemotherapy.

While minimally invasive surgery for adult cancer is accepted as the standard of care for some cancers, including small adrenal tumors, the role of minimally invasive operations in pediatric cancer patients is less well defined. We found that there were decreased numbers of open surgeries in the later cohort for extra-adrenal neuroblastomas, particularly an increased number of minimally invasive surgeries for thoracic primaries. This trend aligns with the overall shifts in the recent years of performing oncologic procedures minimally invasively, with multiple documented patient benefits including reduced post-operative pain and shorter hospital length of stay [ 18]. Recent advances in minimally invasive techniques have increased its utility for both biopsy and resection [ 18]. A 2020 systematic review of MIS approaches in neuroblastoma showed that minimal invasive approaches are safe in small (less than 6 cm) neuroblastomas without IDRFs regardless of primary site [ 21]. These prior study findings are similar to a prospective study in 2014 on adrenal neuroblastoma without IDRF, which demonstrated safety with MIS resection in this patient population [ 19]. More research has been recently performed on the MIS treatment for mediastinal neuroblastoma. A retrospective study in Japan from 2014 to 2016 demonstrated that a thoracoscopic approach to thoracic neuroblastoma is safe and feasible and that tumor size may not limit the ability to perform MIS surgery [ 22]. They additionally performed the same retrospective study looking at abdominal neuroblastoma and concluded that the MIS approach is also safe in abdominal neuroblastoma; the authors suggested a tumor size limit of 6 cm as the upper limit for safe resection for laparoscopy[ 23], though this has not been validated. Overall, the available literature concludes that MIS approaches to neuroblastoma biopsy and resection are safe for small tumors without IDRF for both abdominal and thoracic neuroblastoma.

Surgical resection with curative intent remains a therapeutic goal in the treatment of Neuroblastoma. We hypothesized that with the introduction of the INRG system in 2009 we would see a longer time to surgery in the second decade due to staging no longer being reliant on up front surgical resection [ 13,24]. Among patients with INSS stage 2 and 2b, we did in fact observe a longer time to surgery in the later decade: 3.5 days vs.94 days (

P

= 0.03) and 5.5 days vs. 52 days (

P

< 0.01), respectively. However, in our higher INSS staged patients we saw a shorter time to surgery in the later group, and since most of our patients had stage 4 disease, the average time to surgery for the entire cohort was shorter in the more recent decade (Table 3).While our data set was not complete enough to elucidate this(IDRFs and INRG staging was not reliably reported even in the recent decade), we surmise that this finding is secondary to the fact that a subset of INSS stage 2 tumors, where resection was previously undertaken early to stage tumors are reclassified as INRG stage L2 if IDRFs are present and, therefore, undergo preoperative chemotherapy [ 13]. Possible explanations for the shorter time to definitive surgery overall in the recent decade include: (1) without reliance on open resection for surgical staging, patients underwent core needle or minimally invasive biopsies, completed staging, and started chemotherapy sooner in the second decade than in the earlier INSS era; (2) identification of IDRFs up front and determination of response (or lack thereof)to chemotherapy allowed for preemptive surgical planning; (3)surgical intervention may have been more uniformly timed at a specific timepoints within the chemotherapy treatment protocols in the IDRF era.There are multiple limitations in this study. It is a retrospective study, and we cannot control for differences in the patient populations over the 20-year period. In addition, we had a small sample size during this period and the trends only reflect our single center. Some of our findings during this 20-year period could be explained by generalized practice changes, such as increased comfort with and availability of percutaneous core needle biopsies and minimally invasive surgical resections in children. Conclusions could not be made about the proportion of

MYCN

amplified tumors, as the data was not available for many patients. We did note,however, that a much smaller percent of patients had missing

MYCN

amplification data in the more recent decade.In addition, while we clinically had pediatric radiologists interpreting images performed, there was not an independent review of these images in our study to assure accurate staging. Finally, pathology and operative reporting was not uniform, so some data points were not included.

In conclusion, the trends we saw over our 20-year period are in line with changes in practice patterns globally. We found a transition to more minimally invasive approaches, a shorter time interval to surgery overall, as well as a narrower time window(later for INSS stage 2a, 2b, sooner for INSS stage 4) for surgeries in the course of treatment for neuroblastoma. We cannot conclude that the changes are directly related to the transition to INRG staging. It is more likely that the practice changes over time are multifactorial, and the improved uniformity in surgical timing may also have been attributable to uniformity of oncology treatment protocols. A larger sample size and a multi-institutional study design would better allow us to characterize these findings.

Author contributions

All authors contributed to the study conception and design. SB and PA contributed to the material preparation, data collection and analysis, and wrote the first draft of the manuscript.GCN and CR contributed to review and editing as well as supervision.All authors read and approved the final manuscript.

Funding

No funding was used for this project.

Data availability

The data sets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of interest

No financial or non-financial benefits have been received or will be received from any party related directly or indirectly to the subject of this article. The authors have no conflict of interest to declare.

Ethical approval

This study was approved by the Institutional Review Board (IRB#00014901).