Table 1
Clinical and molecular characterization of the infant patients.
| level | G3 | G4 | SHH | p |
---|
n | | 9 | 3 | 11 | |
Gender (%) | female | 4 (44.44) | 3 (100.00) | 4 (36.36) | 0.1428 |
| male | 5 (55.56) | 0 (0.00) | 7 (63.64) | |
Age (%) | 2 | 2 (22.22) | 0 (0.00) | 2 (18.18) | 0.6762 |
| 3 | 7 (77.78) | 3 (100.00) | 9 (81.82) | |
M_Stage (%) | M+ | 2 (22.22) | 1 (33.33) | 1 (9.09) | 0.5476 |
| M0 | 7 (77.78) | 2 (66.67) | 10 (90.91) | |
Excision (%) | GTR | 7 (77.78) | 1 (33.33) | 7 (63.64) | 0.3711 |
| NTR | 2 (22.22) | 2 (66.67) | 4 (36.36) | |
Treatment (%) | Both | 5 (55.56) | 2 (66.67) | 10 (90.91) | 0.1918 |
| Chemotherapy | 4 (44.44) | 1 (33.33) | 1 (9.09) | |
Recurrence (%) | 0 | 6 (66.67) | 3 (100.00) | 7 (63.64) | 0.4651 |
| 1 | 3 (33.33) | 0 (0.00) | 4 (36.36) | |
Death (%) | 0 | 5 (55.56) | 2 (66.67) | 8 (72.73) | 0.7237 |
| 1 | 4 (44.44) | 1 (33.33) | 3 (27.27) | |
Histology (%) | CMB | 4 (44.44) | 2 (66.67) | 2 (18.18) | 0.3770 |
| DNMB | 1 (11.11) | 1 (33.33) | 6 (54.55) | |
| MBEN | 2 (22.22) | 0 (0.00) | 1 (9.09) | |
| Undefinded | 2 (22.22) | 0 (0.00) | 2 (18.18) | |
OS_TIME (median [IQR]) | | 50.892 [48.888, 66.465] | 72.938 [37.093, 81.316] | 67.254 [41.463, 84.240] | 0.8099 |
A total of 23 infant with medulloblastoma were included in this study, including 12 males and 11 females. 11 samples with pathology of SHH subtype, 9 samples with G3 subtype, and 3 samples with G4 were clustered and classified by tumor samples by methylation sequencing. The follow-up time of all children was 1-131 months, and the median follow-up time was 51 months. The overall survival was 65.22%, the survival rate of SHH was 72.73%, the survival rate of G3 was 55.56%, the survival rate of G4 was 66.67%, and the survival rates of children at 1, 3, 5, and 10 years were 100.00%, 82.61%, 52.17%, and 13.04%, respectively. Clinical data of pediatric medulloblastoma patients of the 23 active cases, age was dichotomized into children aged 1–2 years and 2–3 years, with almost the same proportion of female children as male children. The distribution of molecular subtypes was statistically different among the three groups, with the highest proportion of SHH type 47.83%, G3 type 39.13% and G4 13.04%. This is accord with the typing characteristics of infant medulloblastoma. And in this study, age, sex, histologic typing, treatment, dissemination, recurrence, and survival were not statistically different among the three subtypes (Table 1). In histologic typing, CMB and DNMB accounted for the largest proportions of 69.47% and 34.78% respectively, and in the corresponding molecular typing, the SHH predominantly constituted DNMB and the G3 predominantly constituted CMB. (Fig. 1)
Table 2
Cox Univariate and Multivariate Regression Analysis of Prognostic Factors in Medulloblastoma
| Cox univariate analysis | | Cox multivariate analysis | |
---|
Variants | Hazard Ratio (95%CI) | P-value | Hazard Ratio (95%CI) | P-value |
---|
Diagnosis age | 2.69 (0.43–16.71) | 0.289 | 5.49 (0.44–68.09) | 0.1847 |
Excision: GTR | | | | |
Excision: NTR | 0.56 (0.11–2.75) | 0.472 | | |
Gender: female | | | | |
Gender: male | 0.71 (0.18–2.84) | 0.627 | | |
Histology: CMB | | | | |
Histology: DNMB | 0.17 (0.02–1.72) | 0.133 | | |
Histology: MBEN | 0.72 (0.07–7.04) | 0.781 | | |
Histology: Undefinded | 1.35 (0.26–7.01) | 0.718 | | |
Molectype: G3 | | | | |
Molectype: G4 | 0.7 (0.08–6.45) | 0.752 | | |
Molectype: SHH | 0.54 (0.12–2.45) | 0.423 | | |
MStage: M1 | | | | |
MStage: M0 | 1.66 (0.2-13.72) | 0.638 | | |
Recurrence: 0 | | | | |
Recurrence: 1 | 10.28 (1.99–53.03) | 0.005 | 10.69 (2.04–56.02) | 0.0051 |
Treatment: Both | | | | |
Treatment: Chemotherapy | 4.59 (1.11–18.93) | 0.035 | 4.41 (0.87–22.31) | 0.0726 |
Cox Univariate and Multivariate Regression Analysis
Univariates and multivariate COX risk models were analyzed in this cohort, incorporating factors related to prognosis such as age of Diagnosis, gender, pathology type, molecular type, excision of surgical resection, treatment, whether recurrence, and whether metastasis. Among them, in the univariate analysis, the differences in whether recurrence and treatment had a significant impact on survival (P-values < 0.05), while gender, age of diagnosis, molecular type, and surgical resection were not related to their prognosis (P-values > 0.05) (Table 2). The above meaningful prognostic factors were further analyzed by incorporating them into a Multivariate COX, which showed the same result as Univariates COX. the HR for recurrence (Recurrence: 1) was 10.28, 95% CI was 1.99 to 53.03, with a p-value of 0.005, and in Multivariate COX the HR was 10.69, 95% CI was 2.04 to 56.02, with a p-value of 0.0051, indicating a significant correlation to survival time. Treatment including postoperative chemotherapy alone (Chemotherapy) and postoperative chemotherapy with delayed radiotherapy (Both). The HR for Chemotherapy in Univariates COX analysis was 4.59, 95% CI was 1.11 to 18.93, with p-value was 0.035, and in Multivariate COX the HR was 4.41, 95% CI was 0.87 to 22.31, with p-value was 0.0726. This result suggests that, despite the poor prognosis of medulloblastoma in the less than 3 years of age, aggressive postoperative treatment, i.e., postoperative chemotherapy and delayed radiotherapy, rather than chemotherapy alone, has a greater benefit for overall survival. In this analysis, it is worth noting that in contrast to the results for the all-age group, it is generally believed that the SHH subtype has a poorer prognosis; however, this result was not found in this study.
Survival analysis (Fig. 2) was performed on 23 MBs for overall survival status (a, b) and progression free survival(c, d). Overall survival analysis compared molecular type and treatment. Only the treatment had a difference in the overall survival (p = 0.022), there was no difference between molecular typing(p > 0.05). However, there is no difference among all factors being considered including histology, diagnosis age, extent of resection. The analysis shows that postoperative chemotherapy and delayed radiotherapy harvested a better OS, but not for recurrence. It is worth noting that molecular typing and the extent of resection did not make a significant difference in overall survival, and the extent of resection of MB in children is still controversial. In addition to this, it was noted that children with medulloblastoma with molecular typing of SHH are generally considered to have a worse prognosis, which was not reflected in this study.