Study population

95 newly diagnosed NB patients (male: 46, female: 49) were analyzed in our study. The median age of patients was 2.94 years (interquartile range 1.72–4.73 years). The majority of NB patients (90, 94.7%) had metastases. According to the International Neuroblastoma Staging System (INSS), 1 (1.1%) patient was stage 1, 7 (7.4%) patients were stage 2, 14 (14.7%) patients were stage 3, 73 (76.8%) patients were stage 4. According to Children’s Oncology Group (COG) risk grouping, 6 (6.3%) patients were classified as low-risk, 17 (17.9%) patients as intermediate-risk and 72 (75.8%) patients as high-risk. Ultimately, 90 (94.7%) patients underwent surgery and 93 (97.9%) patients received neoadjuvant chemotherapy. The median follow-up time was 35.0 months (interquartile range 20.5–44.8 months). During clinical follow up, 27 (28.4%) patients died, 21 (22.1%) patients relapsed and 47 (49.5%) patients remained progression-free survival (Table 1).

Table 1 Clinical characteristics of patientsComparing clinicopathological factors and metabolic parameters between the group with and without events

Further analysis was conducted to compare clinicopathological factors and metabolic parameters between the groups with and without events (died or relapsed or progression). There were significant differences in age (P = 0.001), INSS (P < 0.001), COG (P < 0.001), Chromosome 11q (P = 0.004), neuron-specific enolase (NSE, P < 0.001), serum ferritin (SF, P < 0.001), lactate dehydrogenase (LDH, P = 0.002), PSUVmax (P = 0.030), IMH (P < 0.001), WMTV (P < 0.001), WTLG (P < 0.001), WSUVmax (P = 0.001), WSUVpeak (P = 0.007) and WMH (P < 0.001) between two groups. Tumor primary site (P = 0.203), gender (P = 0.123), myelocytomatosis viral oncogene neuroblastoma derived homolog (MYCN, P = 0.652), Chromosome 1p36 (P = 0.696), PMTV (P = 0.072), PTLG (P = 0.123), PSUVmean (P = 0.547), PSUVpeak (P = 0.106) and WSUVmean (P = 0.760) showed no statistically significance between two groups (Table 2).

Table 2 Comparing clinicopathological factors and metabolic parameters between group with and without eventsComparison of 18 F-FDG PET/CT metabolic parameters and tumor metabolic heterogeneity derived from the primary lesion and whole-body lesions

Comprehensive comparisons were performed among different metabolic parameters. In the correlation analysis, PSUVmax and WSUVmax (r = 0.901, P < 0.001), PSUVmean and WSUVmean (r = 0.873, P < 0.001), PSUVpeak and WSUVpeak (r = 0.870, P < 0.001) showed extremely high correlation. Additionally, there was a high correlation between PMTV and WMTV (r = 0.725, P < 0.001), PTLG and WTLG (r = 0.759, P < 0.001), IMH and WMH (r = 0.772, P < 0.001). Our results demonstrated a high positive correlation in metabolic parameters and tumor metabolic heterogeneity between primary lesions and whole-body lesions.

In the differential analysis, no significant difference was observed between PSUVmax and WSUVmax (Z=-1.723, P = 0.085), PSUVmean and WSUVmean (Z=-0.784, P = 0.433), PSUVpeak and WSUVpeak (r=-1.554, P = 0.120). However, there were significant differences between PMTV and WMTV (Z=-4.638, P < 0.001), PTLG and WTLG (Z=-4.258, P < 0.001), IMH and WMH (Z=-2.005, P = 0.045). Our study indicated that the tumor metabolic burden (represented by MTV and TLG) and tumor metabolic heterogeneity were significantly different between primary lesions and whole-body lesions (Table 3).

Table 3 Comparison of metabolic parameters and tumor metabolic heterogeneity derived from primary lesion and whole-body lesionsComparing intra-tumoral metabolic heterogeneity and whole-tumoral metabolic heterogeneity in different subgroups

The differences of IMH and WMH between subgroup of patients were also investigated. NB patients were categorized into three groups according to their INSS staging and COG risk group (non-IV and non-high-risk group, IV or high-risk group, IV and high-risk group). No statistically significant differences were detected in the non-IV and non-high-risk groups (Z=-1.718, P = 0.086). While statistically significant differences were observed in IV or high-risk group (Z=-2.461, P = 0.014), and IV and high-risk group (Z=-3.389, P = 0.001). Our study found that tumor metabolic heterogeneity was significant different in stage IV or high-risk group NB patients (Fig. 2).

Fig. 2

Comparing intra-tumoral metabolic heterogeneity and whole-tumoral metabolic heterogeneity in different subgroups

Survival analysis

Clinicopathological factors and metabolic parameters were included in survival analysis. In univariate analysis, age (P = 0.017), INSS (P = 0.002), COG (P = 0.003), Chromosome 11q (P = 0.007), NSE (P < 0.001), SF (P = 0.003), LDH (P = 0.008), PSUVmax (P = 0.011), PSUVpeak (P = 0.035), PMTV (P = 0.025), IMH (P < 0.001), WSUVmax (P = 0.002), WSUVpeak (P = 0.006), WMTV (P < 0.001), WTLG (P = 0.001), and WMH (P < 0.001) were significantly associated with PFS. Variables with significant differences in the univariate analysis were included in multivariate analysis. Multivariate analysis identified that INSS (P = 0.011), WMTV (P < 0.001), WTLG (P = 0.004) and WMH (P < 0.001) were independent risk factors for PFS (Table 4).

Table 4 Survival analysis of prognostic factors for PFS and OS

In univariate analysis of OS, MYCN (P = 0.025), NSE (P < 0.001), SF (P = 0.006), LDH (P < 0.001), PSUVmax (P = 0.001), PSUVpeak (P = 0.003), IMH (P = 0.003), WSUVmax (P < 0.001), WSUVpeak (P = 0.001), WMTV (P = 0.003), WTLG (P = 0.012), and WMH (P < 0.001) were significant associated with OS. However, in multivariate analysis, only NSE (P = 0.002) and WMH (P < 0.001) remained significant. Therefore, NSE and WMH were identified as independent prognostic risk factors for OS (Table 4).

Further evaluate tumor metabolic burden and WMH on survival

Based on the results of multivariate survival analysis, we further investigated the predictive effect of whole-body tumor metabolic burden and WHM. The optimal cut-off values for WMTV, WTLG and WMH were 413.14, 1044.14 and 0.448, respectively, determined by the area under the receiver operating characteristic curve. Significant differences were observed in WMTV and WTLG for PFS (P < 0.001, P < 0.001, respectively) and OS (P < 0.001, P < 0.001, respectively). Similarly, WMH also manifested significant differences for PFS (P < 0.001) and OS (P = 0.004) (Fig. 3). Two representative NB patients with high and low WMH values were presented in Fig. 4.

Fig. 3

Survival curves based on optimal cut-off value

Fig. 4

Two representative NB patients with high and low WMH values; A: a 32-month-old girl with a relatively uniform FDG uptake in whole-body tumor (White arrow, IMH:0.443, WMH:0.470), was disease free survival 51.3 months after diagnosis; B: a 29-month-old girl with a heterogeneous FDG uptake in whole-body tumor (Red arrow, IMH:0.463, WMH:0.392), relapsed at 7.1 months, and died at 8.9 months after diagnosis