Clinical and Pathological Evidence of Anti-GD2 Immunotherapy Induced Differentiation in Relapsed/Refractory High-Risk Neuroblastoma

Abstract

Background: Neuroblastic tumors (NBTs) originate from a block in the process of differentiation. Histologically, NBTs are classified in neuroblastoma (NB), ganglioneuroblastoma (GNB), and ganglioneuroma (GN). Current therapy for high-risk (HR) NB includes chemotherapy, surgery, radiotherapy, and anti-GD2 monoclonal antibodies (mAbs). Anti-GD2 mAbs induce immunological cytoxicity but also direct cell death.

Methods: We report on patients treated with naxitamab for chemorefractory NB showing lesions with long periods of stable disease. Target lesions with persisting ¹²³I-Metaiodobenzylguanidine (MIBG) uptake after 4 cycles of immunotherapy were further evaluated by functional Magnetic Resonance Imaging (MRI) and/or Fluorodeoxyglucose (FDG)-positron emission tomography (PET). MIBG avid lesions that became non-restrictive on MRI (apparent diffusion coefficient (ADC) > 1) and/or FDG-PET negative (SUV < 2) were biopsied.

Results: Twenty-seven relapse/refractory (R/R) HR-NB patients were enrolled on protocol Ymabs 201. Two (7.5%) of the 27 showed persistent bone lesions on MIBG, ADC high, and/or FDG-PET negative. Forty-four R/R HR-NB patients received chemo-immunotherapy. Twelve (27%) of the 44 developed persistent MIBG+ but FDG-PET- and/or high ADC lesions. Twelve (86%) of the 14 cases identified were successfully biopsied producing 16 evaluable samples. Histology showed ganglioneuroma maturing subtype in 6 (37.5%); ganglioneuroma mature subtype with no neuroblastic component in 4 (25%); differentiating NB with no Schwannian stroma in 5 (31%); and undifferentiated NB without Schwannian stroma in one (6%). Overall, 10 (62.5%) of the 16 specimens were histopathologically fully mature NBTs.

Conclusions: Our results disclose an undescribed mechanism of action for naxitamab and highlight the limitations of conventional imaging in the evaluation of anti-GD2 immunotherapy clinical efficacy for HR-NB.

Keywords: CHD5; anti-GD2 immunotherapy; differentiation; functional MRI; functional imaging; naxitamab; neuroblastoma.

Figure 1

ADC in MRI+ observations.

Figure 2

Correlation example of positive bone lesions by ¹²³I-Metaiodobenzylguanidine (MIBG) and diffusion-weighted (DW) Magnetic Resonance Imaging (MRI) for patient #9 (Table 1) before chemo-immunotherapy treatment.

Figure 3

Example of a good correlation between proximal left femoral metaphysis ¹²³I-Metaiodobenzylguanidine (MIBG) bone uptake (left column) and MRI/DW (middle and right panels) for patient #13 (Table 1), showing T1 hypointense, STIR T2 hyperintense focal bone lesion with restricted diffusion (b 1000), and low ADC (<1.01) values.

Figure 4

Correlation of functional imaging for patient #1 (Table 1) and histopathology. (A) Correlation between pre-immunotherapy treatment (top panels) functional imaging of chemo refractory iliac bone lesion and post immunotherapy evaluation (bottom panels: STIR T2 hyperintense focal iliac bone lesion with restricted diffusion -b 1000- and high- > 1.01- ADC values) of the same iliac lesion. (B) Bone tumor from patient #1 post immunotherapy. This sample shows ganglioneuromatous tissue with Schwannian stroma and some mature ganglion cells (arrow). This sample corresponds to the iliac bone lesion described by functional imaging in (A) (bottom panels). (C) S100 immunostaining highlighting the Schwannian stroma (arrow) of the same iliac bone tumor.

Figure 5

Describes case #4 (Table 1 and Table 2), a patient managed initially with naxitamab and GM-CSF with no objective response (MIBG and PET-FDG diffusely positive in multiple bone lesions). Subsequently patient received HITS showing an objective response (lower Curie score) and PET-FDG turning negative. (A) Bone tissue diffusely infiltrated by tumor composed of Schwannian stroma (arrow), (B) S-100 immunostaining depicting massive Schwannian stroma (arrow).

Figure 5

Describes case #4 (Table 1 and Table 2), a patient managed initially with naxitamab and GM-CSF with no objective response (MIBG and PET-FDG diffusely positive in multiple bone lesions). Subsequently patient received HITS showing an objective response (lower Curie score) and PET-FDG turning negative. (A) Bone tissue diffusely infiltrated by tumor composed of Schwannian stroma (arrow), (B) S-100 immunostaining depicting massive Schwannian stroma (arrow).

Figure 6

Describes case #5 (Table 1 and Table 2) biopsy of the right sphenoid bone. (A) Histology consists of only Schwann tissue (arrow) with no evidence of neuroblastic cells. (B) Schwannian stroma shown by S100 immunostaining.