Management of Primary Thoracic Neuroblastic Tumors in Children: An Observational Study.

Aims: The mediastinum is the second common site of origin of pediatric neuroblastic tumors. Primary thoracic neuroblastoma (NB) is considered to be a site with favorable outcome and is reported to have a better prognosis than the other sites of origin. This is an observational study on our experience in the management of children with primary thoracic neuroblastic tumors. Materials and
Methods: A retrospective observational review of the medical records of all the children treated for primary thoracic neuroblastic tumors including NB, ganglioneuroblastoma, and ganglioneuroma over a period of 8 years from January 2011 to December 2018 at our Institute was performed. We analyzed the factors including age, stage, histology, symptoms at presentation, surgical management, adjuvant treatment, and the 2-year survival of patients.
Results: A total of 23 cases of primary thoracic neuroblastic tumors were identified during the study period. Ten patients had Stage III disease (43%), six had Stage IV (26%), four had Stage II, and three patients had Stage I. A total of four children (17%) died during the study period. A total of 13 children underwent complete excision of the tumor.
Conclusion: In this retrospective review, we have observed that the mediastinal site of origin alone did not uniformly confer an excellent prognosis for all the patients. While the patients with lower stage tumors, favorable biological profile, and infants had excellent prognosis, in another small subset of patients with undifferentiated histology, the prognosis was guarded with an increased risk of recurrence. The clinical presentation with pleural effusion and compression of the airway at the time of presentation was associated with poor outcome. Copyright:

INTRODUCTION

The mediastinum is traditionally considered to be a favorable site of origin for pediatric neuroblastic tumors with a better outcome.[1] The biologic variables such as amplification of N-myc, DNA ploidy, serum ferritin levels, and serum lactate dehydrogenase levels were considered to predict the prognosis of the patients along with the age and stage at presentation. Morris et al. after reviewing the biologic variables had concluded that the thoracic site of origin was an independent predictor for better prognosis in children with neuroblastoma (NB).[2] In this observational study, we reviewed our institutional experience in the management of primary thoracic neuroblastic tumors, the clinical presentation, and 2-year survival.

MATERIALS AND METHODS

A retrospective review of all children treated for primary thoracic neuroblastic tumors over an 8-year period between January 2011 and December 2018 at the pediatric surgery department in our institute was performed. Children with biopsy-proven NB, ganglioneuroblastoma (GNB), and ganglioneuroma were included in the study. Children with other mediastinal tumors such as germ cell tumor and Ewing's sarcoma were excluded from the study.

The medical records were reviewed for patient demographics (age at presentation and sex), presenting symptoms, preoperative staging evaluation (24-h urinary vanillylmandelic acid [VMA] level, bone marrow biopsy, contrast computed tomography of the chest for primary tumor evaluation, magnetic resonance imaging of the spine for spinal extension, and Technetium 99 m bone scan for bony metastasis), preoperative treatment (chemotherapy), operative details (open thoracotomy or Video-assisted thoracic surgery (VATS) excision or biopsy alone), tumor histology (NB, GNB, or ganglioneuroma), and outcome (2-year survival, early and late complications, recurrence, and mortality).

The indications for upfront surgery with primary excision included children with localized tumors <5 cm, not involving vital structures as defined by the list of image-defined risk factors (IDRF) in the preoperative imaging, and no evidence of metastasis on evaluation. The IDRF criteria included tumors encasing aorta and/or major branches, tumors compressing the trachea and/or principal bronchi, lower mediastinal tumors infiltrating the costovertebral junction between T9 and T12. Children with larger tumors (more than 5 cm) with the presence of IDRF were subjected to initial biopsy followed by neoadjuvant chemotherapy. This was followed by the excision of the tumor depending on the response to the chemotherapy. The decision on the chemotherapy regimen and the need for radiotherapy were discussed and approved by the institutional tumor board. Smaller tumors were excised using video-assisted thoracoscopic surgery (VATS) and larger tumors were excised by posterolateral thoracotomy. The tumors were staged according to the International Neuroblastoma Staging System after the staging investigations and attempted excision.

Chemotherapy was advised by the pediatric oncologist in the institutional tumor board. The three-drug regimen comprised cisplatin (80 mg/m2), vincristine (1.5 mg/m2), and cyclophosphamide (1000 mg/m2). The duration of chemotherapy depended on the risk stratification of the disease with low-risk category receiving fewer cycles and high-risk category patients typically receiving 3-weekly chemotherapy for 10–12 cycles. Etoposide, ifosfamide, and adriamycin were used in select high-risk group patients with high tumor burden or as part of salvage chemotherapy for patients with recurrence.

Radiotherapy was given as an adjuvant therapy for select patients who did not undergo complete excision due to persistence of IDRF for high-risk patients who underwent partial excision with significant macroscopic residue, and for patients who had developed recurrence. The standard dosage used was 21.6 Gy in 1.8 Gy fractions. This is an observational study aimed at elaborating the clinical presentation and management of primary thoracic neuroblastic tumors at our institute and we have used simple descriptive statistics to analyze the variables in our study.

RESULTS

A total of 23 cases of primary thoracic NB were identified during the study period at our institution between January 2011 and December 2018. A total of 15 were boys (65%) and eight of them were girls (35%). The age at presentation was varied with four children (17%) presenting during the infancy, six children (26%) presenting between the 1 and 2 years of age, five of them (22%) between 2 and 5 years of age, and eight children (35%) over the period of 5 years and <12 years. The age of presentation in our series ranged from 6 months to 11 years of age with the interquartile range of 54 months. A total of 12 patients (52%) had left-sided thoracic mass and the remaining 11 (48%) had right-sided thoracic mass [Table 1].

Table 1

Summary of the staging and management of all the patients in the study

Age/sex	Side	Spinal extension	INSS	VMA/BMA/N-Myc	Surgery timing/type	HPE	CT	RT	Recurrence	2 years survival
6/male	Left		II A	Negative/negative	Excision after CT	GNB	Yes	No	No	Yes
1.5/male	Left		III	Negative/negative	Biopsy alone	NB	Yes	No	No	Yes
7/female	Right	Yes	III	Negative/negative	Excision after CT	NB	Yes	Yes	Yes	Yes
11/12/male	Right		I	Negative/negative	Primary excision	GNB	No	No	No	Yes
10/12/female	Left		IV	Negative/positive	Biopsy alone	NB	Yes	No	No	No
6/12/male	Right		IV	Positive/positive	Biopsy alone	NB	Yes	No	No	Yes
3/male	Right	Yes	III	Negative/negative	Biopsy alone	NB	Yes	No	No	Yes
4/female	Right		IV	Positive/negative	Biopsy alone	NB	Yes	No	No	Yes
11/female	Left		III	Negative/negative	Primary excision	GN	No	No	No	Yes
3/male	Left		III	Negative/negative	Biopsy alone	NB	Yes	No	No	Yes
6/male	Left		IIA	Negative/negative	VATS excision after CT	NB	Yes	No	No	Yes
1.5/male	Left	Yes	III	Negative/negative	laminectomy/VATS after chemo	NB	Yes	No	No	Yes
2/female	Left		III	Positive/negative	Excision after CT	NB	Yes	No	-	No
10/male	Left		II	Negative/negative	Primary excision	GNB	Yes	No	Yes	Yes
6/12/male	Right		I	Negative/negative	Primary excision	GNB	Yes	No	No	Yes
4/female	Right		I	Negative/negative	Primary VATS excision	GNB	Yes	No	No	Yes
2.5/male	Left	Yes	III	Negative/negative/N-Myc −	Partial excision after CT	NB	Yes	Yes	No	Yes
11/female	Left		IV	Negative/positive/N-Myc−	Biopsy alone	NB	Yes	No	No	Yes
6/female	Right		IV	Negative/positive/N-Myc−	Biopsy alone	NB	Yes	Yes	Yes	Yes
11/male	Left		III	Positive/negative/N-Myc−	Biopsy alone	NB	Yes	No	No	Yes
3.5/male	Right		III	Negative/negative/N-Myc −	Excision after CT	NB	Yes	No	Yes	No
1.5/male	Right		IV	Positive/positive/N-Myc +	Excision after CT	NB	Yes	No	No	No
1.5/male	Right		II	Negative/negative/N-Myc −	Primary excision	NB	Yes	No	No	Yes

INSS: International neuroblastoma staging system, VMA: Vinillyl mandelic acid, BMA: Bone marrow aspiration, CT: Chemotherapy, RT: Radiotherapy, NB: Neuroblastoma, GNB: Ganglioneuroblastoma, GN: Ganglioneuroma, N-Myc: +/−amplification/not amplified, VATS: Video-assisted thoracic surgery, HPE: Histopathology

The common presenting symptoms included respiratory distress, upper respiratory symptoms, fever, constant chest pain, opsoclonus-myoclonus syndrome, paraparesis, horner's syndrome, and neck node. Five children had an incidental diagnosis of thoracic mass during a screening chest X-ray taken to rule out pneumonia. The child with opsoclonus-myoclonus syndrome presented with nystagmus and was diagnosed with a mediastinal mass on screening. The patient with horner's syndrome at presentation was diagnosed with right mediastinal mass near the thoracic inlet.

The 24-h urinary VMA levels were elevated in five patients (22%). Three of them had stage IV disease and two patients had Stage III disease. VMA was not elevated in Stage 1 and stage 2 disease. The bone marrow was positive for tumor cells in five patients (22%) with Stage IV disease [Table 1].

A total of four patients had intraspinal extension of the tumor and two of them presented with paraparesis. One patient underwent urgent laminectomy and excision of the extradural mass. The paraparesis recovered after the surgery. A patient with long duration paraparesis had partial recovery after the chemotherapy-induced disappearance of the intraspinal component. The remaining two patients with intraspinal tumor extension without neurological deficit underwent complete excision of the extraspinal component after neoadjuvant chemotherapy. The intraspinal tumor component showed complete response to the chemotherapy. The primary lesion was NB in all four patients [Table 1].

The primary excision of the tumor was done in six patients (26%). The remaining patients (n = 17) underwent an initial biopsy followed by neoadjuvant chemotherapy. Four of them had Stage IV disease at presentation and the remaining 13 patients were Stage III. Seven of these 13 Stage III patients underwent complete excision after the neoadjuvant chemotherapy. Hence, a total of 13 patients (57%-Stage I to III) underwent complete tumor excision [Table 1].

A total of ten patients (43%) did not undergo complete tumor excision. Two of them were sick at presentation and died shortly after instituting neoadjuvant chemotherapy. Of the eight remaining patients, four patients had Stage III disease and had persisting IDRF after the neoadjuvant chemotherapy and hence were not operated. Chemotherapy was continued. Three children had complete response at the end of the adjuvant chemotherapy and are on follow-up without recurrence. One patient had a small residual mass encircling the descending thoracic aorta at the completion of adjuvant treatment and received 13-cis-retinoic acid for six cycles. The small residual mass remained inactive and stable on the 4th year of follow-up.

Of the remaining four patients with Stage IV disease, three of them were positive for bone marrow metastasis and one with skull metastasis at presentation. All four of them did not undergo primary tumor excision and received adjuvant chemotherapy alone. At the end of the chemotherapy, these four Stage IV patients showed complete response. However, one patient developed nodal recurrence on follow-up and received salvage radiotherapy and chemotherapy.

A total of 13 patients underwent complete tumor excision with or without microscopic residue (primary excision in six patients and after neoadjuvant chemotherapy in seven patients). Ten of them underwent open excision through a posterolateral thoracotomy. Two cases who underwent open excision had major vascular infiltration with significant blood loss during the surgery and required vascular repair. A total of three patients underwent thoracoscopic (VATS) excision of the tumor. One patient who underwent VATS excision had 180°degree adhesion to the aorta and despite careful dissection during VATS, an inadvertent aortic injury occurred. The procedure was converted to an open thoracotomy and the aortic injury was repaired.

Stage III (n = 10) and Stage IV (n = 6) patients constituted about 70% (16 patients) of the total cases in this study. The remaining 30% (seven patients) included the Stage I (n = 3) and Stage II (n = 4) patients [Table 1]. All the Stage I and II patients underwent complete excision and are surviving. One Stage II patient had nodal recurrence on follow-up.

A total of eight of the ten Stage III patients are surviving and two patients had died. Four of the eight Stage III patients underwent complete excision and four patients did not undergo excision. One Stage III patient who underwent complete excision had a recurrence. None of the four patients who did not undergo complete excision develop recurrence.

A total of six patients had Stage IV disease and five patients had bone marrow metastasis and one patient had skull metastasis. Two children died and four of them are surviving. Of the four patients, none underwent complete tumor excision and three of them are disease-free at the end of adjuvant chemotherapy. One patient developed recurrence and received salvage chemo-radiotherapy and is surviving.

The histopathology was NB in 17 patients (74%), GNB in five patients (22%), and Ganglioneuroma (GN) in one patient (4%) [Table 1]. Of the 17 patients with NB, five patients had undifferentiated NB and the remaining had either poorly differentiated or differentiating type. The 2-year survival of GNB and GN patients was 100%. A total of four patients died in the NB group with a 2-year survival of 77% (13 out of the 17 patients). Three patients who developed recurrence had undifferentiated NB.

Amplification of N-myc was done in seven patients due to the availability in the last 2 years of the study period. One patient had amplification of N-myc and had an aggressive tumor with airway compression, rib erosion, and intraspinal extension. The tumor did not respond to neoadjuvant chemotherapy and the patient succumbed to the disease. The remaining six patients did not reveal N-myc amplification and they survived.

Recurrence occurred in four patients (13%) who had survived a year after the diagnosis. One Stage III patient who underwent complete tumor excision after neoadjuvant chemotherapy and finished a year of adjuvant chemotherapy presented with multiple lesions at the primary site with pleural effusion after 2-years of disease-free period. The patient rapidly succumbed despite salvage chemotherapy. The other three patients who developed recurrence had Stage II, III, and Stage IV disease. The Stage II and Stage III patients had undergone complete tumor excision and adjuvant chemotherapy and were disease-free for a period of 3 and 9 months, respectively, before being diagnosed with a recurrence. The Stage III patient presented with horner's syndrome at presentation had recurrence of the tumor after complete resection. Both underwent salvage chemotherapy and radiotherapy and are now disease-free. The Stage IV patient who developed recurrent disease received salvage chemo-radiotherapy.

Of the four patients who died during the study period, two patients had pleural effusion and respiratory distress at presentation and they died soon after the diagnosis and initiation of neoadjuvant chemotherapy. One child had a large mass with severe tracheal compression and mediastinal displacement that persisted after neoadjuvant chemotherapy and had prolonged postoperative ventilation after surgery and the child eventually died of ventilator-associated pneumonia. The fourth child who died had developed local recurrence with pleural effusion after a 2-year disease-free period.

DISCUSSION

On reviewing similar pediatric thoracic NB patient series published in the literature,[2345] we observed that some variables were slightly different in our study probably due to demography and inherent nature of the disease in Indian patients. Seventeen percent of patients were <1-year of age in our study in contrast to most series where the presentation at infancy predominated.[4] Filler et al.[1] and Morris et al.[2] reported similar death rate (14%–18%) and survival (82%–85%) as in our study, while Horiuchi et al.,[4] Malek et al.,[5] reported 100% survival.

For patients with primary thoracic NB with intraspinal extension presenting with neurological deficit, an urgent laminectomy is indicated. In our series, since chemotherapy resulted in disappearance of the intraspinal tumor component in three patients, chemotherapy might be effective in patients without neurological deficit. However, long-term follow-up is needed to validate this observation.

All the children who presented with respiratory distress in this study died suggesting that respiratory distress due to pleural effusion and airway compression is a symptom of advanced disease and poor survival in our study.

Primary excision of the tumor was possible in one-fourth of the patients. Small tumors without IDRF in imaging were excised without complications. Complete excision of the tumor was done only in 57% of patients. The IDRF help identify the candidates for primary resection. Aggressive complete resection is not recommended as the prognosis is not significantly altered in the presence of macroscopic or microscopic residual disease.[4] Conservation of normal anatomy is as important as tumor clearance and the objective should be maximum resection with minimal damage to the native structures.[6]

VATS excision was used in two cases in this study and was found to be comparable to open thoracotomy with respect to the tumor clearance and complications.[7] However, more cases and long-term follow-up is needed to validate this observation. Care should be taken to prevent vascular injuries during open or minimally invasive excision of tumors that are adherent to major vascular structures. Chemotherapy is a valuable adjuvant treatment that can achieve complete remission in tumors with persistent IDRF. Radiotherapy can be used in patients with persistent IDRF after neoadjuvant chemotherapy.

The patients with inoperable tumors due to encasement of major vascular structures, but had achieved complete response with chemo-radiotherapy alone had remained disease-free during the study period. This subset of patients had excellent prognosis among mediastinal NB patients.[8] Surprisingly, all the four patients who developed recurrence had undergone complete resection signifying that recurrence is not directly related to incomplete excision.

There are a few limitations in this study because of the inherent resource challenges in the tertiary care center that resulted in the incomplete evaluation of the biological profile of the tumors. Many significant variables such as DNA ploidy status, TRK A and B expression, and N-myc amplification could not be studied completely in all the study patients. This restricted the understanding of the role of these important biological variables in the prognosis of the patients in the study group. There were instances of irregular follow-up in patients from distant places after 2 years. Nevertheless, the available factors still provide a better insight on the behavior of the primary thoracic NB.

CONCLUSION

The mortality rate and recurrence rate of our study show that the mediastinal site of origin alone did not uniformly confer an excellent prognosis for all the patients. The subset of patients with lower stage, favorable histology, and biological factors such as nonamplification of N-myc had better 2-year survival. The 2-year survival with good quality of life is observed in 15 cases (66%). VATS excision is possible for excision for localized tumors without IDRF. Adequate precautions should be taken during the excision of tumors either during open or thoracoscopic approach to prevent catastrophic vascular injuries. The presentation with pleural effusion, airway compression, and respiratory distress resulted in poor outcome.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.