In this study, we investigated the clinical outcomes and prognostic factors, including clinical, radiologic, and pathologic factors in patients under 30 years of age who received radiotherapy for DIPG in Korea. As reported in previous literature [11], the number of brainstem glioma patients per institute was limited due to its rarity. Thus, we designed this multicenter study including 10 tertiary academic institutions for about 20 years. In our study, we found that the median survival time was 13 months and the percentage of long-term survivors was 16.7%, which was similar to other previously reported results [3, 8, 12].
As with previous studies, age ≥ 10 years and good performance status at diagnosis were confirmed as prognostic factors in our study [1, 3]. Furthermore, salvage treatment performed in these patients resulted in better prognosis. Notably, the recently treated patients had significantly better survival than those treated before the year of 2010, even after adjusting for the other clinicopathological factors, imaging features, and treatments. As the pattern of practice and the dose of radiotherapy for DIPG remained unchanged throughout the study period, the improvement in survival may be driven by the evolution of general management for patients and the active intervention for the problem that the patients encountered, probably due to an improved socio-economic condition with good medical insurance reimbursement. Using Surveillance, Epidemiology and End Results (SEER) data, Brandel et al. reported that the survival of grade 3 oligodendroglioma patients improved over time even after adjusting treatment. They also suggested that the survival improvement may be owing to the evolving patterns of medical management [13].
Historically, biopsy has not been recommended for diagnosis, and surgical resection was hindered by the location and infiltrative nature of the tumor. This trend was reconfirmed through a recent survey from the European Society for Pediatric Oncology (SIOPE) brain tumor group [14]. Recent advances in the stereotactic surgery of the brain have enabled us to obtain tissue for genomic analyses, and genetic studies are being conducted accordingly. It was found that the H3K27M mutation exist in approximately 80% of DIPGs [6, 15], and a new disease entity “diffuse midline glioma, H3K27M-mutant” has been incorporated in the 2016 WHO classification of central nervous system tumors [5]. However, genetic or histologic findings have not yet led to a meaningful therapeutic change in the real practice [16]. In our study, biopsy was performed in 18.5% of patients, and surgical resection was done only in 8.7% of patients. Moreover, molecular parameters were obtained only from a very limited number of patients, so it was impossible to conduct further analysis on the molecular parameters. Therefore, many studies regarding the prediction of the prognosis of patients with DIPG were based on imaging findings, which may be applied easily in the actual clinical environment [17–19].
We also investigated the MRI features associated with OS in this study. In the univariate analysis, extrapontine extension of tumor, cystic/necrotic feature, and post-radiotherapy necrosis were prognostic factors related to OS; of which, only post-radiotherapy necrosis was a significant prognostic imaging feature for survival in the multivariate analysis. Recently published report of 357 patients from the international DIPG registry showed that tumor extension beyond pons, enhancement, and tumor necrosis were poor prognostic imaging features related to OS in univariate analysis, but no imaging feature was significant in multivariate analysis [18]. Extrapontine extension was a poor prognostic factor for OS in their study, while it was associated with good prognosis in our study. Interpreting this result, we hypothesized that the rapid growth of tumor was related with the development of symptoms before the development of extrapontine extension, whereas less aggressive tumors were likely diagnosed late with extrapontine extension due to its late onset of symptoms. Additionally, we observed that patients with extrapontine extension had less cystic or necrotic features than those without (32.3% vs. 54.1%, P = 0.016), which may suggest more indolent features of tumors with extrapontine extension. Nonetheless, considering that it was not statistically significant when adjusting patient’s demographics and treatments, further studies investigating MRI features are needed.
Despite the efficacy of bevacizumab in adult glioblastoma, little is known about its efficacy in pediatric patients with newly diagnosed or recurrent DIPG [20–22]. The multinational collaborator study by Hoffman et al. reported that the use of bevacizumab at diagnosis showed greater odds of long-term survival in multivariable logistic regression [1]. Recently, Crotty et al. reported a single-center experience of a 3-drug maintenance regimen of TMZ, irinotecan, and bevacizumab following radiotherapy, which showed prolonged survival in patients with DIPG compared to historical single-agent TMZ [22]. In our study, bevacizumab was used in 23 patients. The OS was better in patients treated with bevacizumab than in those who were not. When patients were divided according to the post-radiotherapy necrosis, we found that the difference in survival was more pronounced in patients with post-radiotherapy necrosis than in those without, although it was not significant. In light of these results, given the small number of patients of our study, we may suggest that the use of bevacizumab in pediatric DIPG patients could be beneficial, especially in patients with increased post-radiotherapy necrosis. Further well-designed trials are needed to determine the potential efficacy of bevacizumab in pediatric DIPG patients.
All patients included in this study underwent radiotherapy as their initial treatment, so we intended to analyze whether the radiation dose and target volume had an effect on the prognosis. However, it was difficult to analyze the effect of the radiotherapy dose on survival because most patients received a relatively homogenous dose of radiation (median 54 Gy) despite the data collected by several clinicians from multiple institutions covering a relatively long period of 19 years. Considering that most failures were infield progression, and the pattern of failures were not affected by the radiotherapy volume, we suggest that a large margin of 1 cm or more to the target volume may not be beneficial. Similar results from 97 patients were reported by Tinkle et al. [23]. They also concluded that no apparent survival or tumor-control benefit was achieved by extending the CTV margins beyond 1 cm.
Several studies have been conducted on neoadjuvant, concurrent, and adjuvant chemotherapy in addition to radiotherapy to improve the prognosis of DIPG patients [24–26]. A phase II study evaluating the efficacy of chemoradiotherapy with TMZ followed by adjuvant TMZ conducted by the Children's Oncology Group showed a disappointing result [9]. In their study, the 1-year event-free survival (EFS) rate was 14%, which failed to achieve a higher rate than the historical baseline of 21.9% observed in CCG-9941. Although these disappointing results on TMZ have been reported by several prospective studies, we found that concurrent or adjuvant use of TMZ was still carried out in our practice depending on the clinician's discretion. This suggests that there are still no applicable drugs that are significantly effective to DIPG patients. In our study, we also could not find any beneficial effect of TMZ as a concurrent or adjuvant therapy after radiotherapy.
Our study had limitations. First, there was an unavoidable bias due to its retrospective nature. Considering the low incidence of DIPG, it is nearly impossible to recruit a large number of patients in the prospective setting. To minimize the bias from the retrospective design, we strictly applied a detailed eligibility criteria and tried to include a homogenous patient group. Second, the sample size was relatively small despite pooling patients treated at 10 tertiary institutions over a period of 19 years. However, because there was no significant change in the treatment policy and radiotherapy dose during the study period, the attributable bias might be minimal. Third, there was very limited information on molecular findings to proceed with the analysis. We thought that it reflected the actual clinical situation where pathologic confirmation was not usually performed unless in a trial setting.
In conclusion, radiotherapy, as a mainstay of treatment for pediatric DIPG, was performed without significant change in specifics such as dose scheme and target volume. The prognosis of DIPG patients was poor despite slight improvement in survival over time. As radiation necrosis was a prognostic factor related to OS, efforts are needed to reduce occurrence of post-radiotherapy necrosis. Furthermore, the use of bevacizumab for radiation necrosis may be helpful in some patients. Future prospective studies to elucidate its role will be warranted.