This study provides evidence that palliative re-irradiation in children and young adults with DIPG who experience progression after initial RT may enhance OS and is associated with a significant rate of symptomatic improvement.

Despite advancements in RT technologies, new systemic therapies, and a better understanding of molecular pathogenesis, the prognosis of DIPG has not significantly improved over time and remains a leading cause of mortality in pediatric and young adult patients, even in the modern era [5, 6]. This underscores the urgent need for research into universally applicable treatment approaches. Given the tumor’s complex location and the limited role of surgery, RT remains the standard treatment. Despite debates and studies indicating no survival advantage over RT alone, chemotherapeutic agents continue to be frequently combined with RT in routine clinical practice [7,8,9]. In our study, the 1‑year PFS rate was found to be higher in patients who received concurrent temozolomide with RT; however, this benefit did not translate into an improvement in OS. Given that MGMT status is well established in adult glial tumor studies as a key predictor of temozolomide efficacy, it was hypothesized that patients undergoing concurrent temozolomide treatment might have harbored MGMT-methylated tumors [10]. However, since the MGMT status of patients in our study is unknown, the observed PFS benefit associated with temozolomide cannot be conclusively interpreted, and it should be noted that other parameters may also have contributed to this finding.

In nearly all patients diagnosed with DIPG, disease progression is ultimately inevitable following RT ± chemotherapy, with limited treatment options available following progression. These options include best supportive care, second-line systemic therapies, and re-irradiation. ONC201 is a promising pharmacotherapeutic agent, particularly for tumors such as DIPG with H3K27M histone mutations, especially in recurrent disease [11]. Despite ongoing clinical trials, it has yet to be integrated into standard treatment protocols, and its global accessibility and practical application remain highly limited. On the other hand, re-irradiation remains an accessible and practical treatment option for this patient group with a particularly poor prognosis. However, the rapid progression that often follows initial RT restricts the doses that can be delivered during re-irradiation to palliative levels. Despite the limited patient numbers, several retrospective studies and two small prospective studies indicate that re-irradiation can provide clinical benefits, even with these palliative doses (Table 4; [12,13,14,15,16,17,18,19,20,21,22,23,24,25,26]). A Canadian retrospective study including 16 patients administered re-irradiation doses of 21.6–36 Gy, achieving a median OS of 6.5 months after progression [17]. This study also demonstrated improved OS compared to a historical cohort that did not receive re-irradiation. A single-center retrospective study conducted at Tata Memorial Hospital examined 20 patients who underwent response-based re-irradiation with relatively high total doses (33.8–43.2 Gy) by literature standards, reporting a median OS of 5.5 months after re-irradiation [18]. The SIOP-E-HGG/DIPG working groups’ retrospective study compared 31 patients who received re-irradiation with 39 who did not [20]. This matched analysis showed that re-irradiation increased OS by 3.4 months post-progression and improved neurological symptoms in nearly 80% of patients. In a subsequent re-analysis, the authors found that re-irradiation doses of ≥ 20 Gy were associated with more significant improvements in the clinical response of ataxia [27]. In a systematic review and meta-analysis involving 90 patients, clinical improvement and radiologic response rates following re-irradiation were reported as 87% and 69%, respectively [28]. A prospective phase I/II trial involving 12 patients employed three different re-irradiation schedules: 24, 26.4, and 30.8 Gy in 12, 12, and 14 fractions, respectively [16]. The authors concluded that re-irradiation can safely be delivered for progressive DIPG and clinical improvement was seen in almost all patients. In our study, re-irradiation administered following disease progression was associated with an approximately 3‑month increase in median OS and a substantial improvement in neurological symptoms. These findings corroborate existing literature, which suggests that re-irradiation, despite being constrained to palliative dose levels, can achieve notable clinical benefits.

While the benefits of re-irradiation for patients with progressive DIPG are well studied in the literature, the optimal protocol for re-irradiation remains unclear, and there is considerable variability among different centers. Alongside conventional RT delivered in < 3 Gy per fraction, there are also data available on hypofractionated re-irradiation protocols using doses greater than 3 Gy per fraction. Although achieving prompt palliation is highly desirable for this patient group with limited life expectancy, it is essential to carefully manage the balance between potential benefits and the increased risk of toxicity associated with higher fractional doses and cumulative brainstem doses. In a recent retrospective study conducted at the Memorial Sloan Kettering Cancer Center, 14 out of 20 patients received a re-irradiation dose of 30–36 Gy in 3 Gy per fraction [29]. The authors observed a reduction in steroid use and clinical improvement in most patients, with no evidence of radiation necrosis. They also noted that survival outcomes with 3 Gy per fraction align with those reported in the literature, suggesting that hypofractionated re-irradiation may be a safe and effective treatment option. However, in this patient group, where survival typically does not exceed 6 months even after re-irradiation, it should be noted that the necessary duration for the development of radiation necrosis is often not reached. Additionally, in the Canadian study, one patient who received 30 Gy of re-irradiation in 10 fractions developed pontine necrosis, but the article does not specify when it occurred [17]. In the previously mentioned study from Tata Memorial, for patients who achieved a clinical response with re-irradiation doses ranging from 21.6 to 30.6 Gy, the total dose was escalated to levels of 39–45 Gy [18]. However, among the 13 patients who received re-irradiation doses > 30.6 Gy, two (15%) experienced grade 5 sudden intratumoral hemorrhage—one following a dose of 45 Gy and the other 43.2 Gy. Therefore, although radiation necrosis may be clinically less significant in this patient group with limited life expectancy, further studies are needed to assess the risk of intratumoral hemorrhage associated with increased cumulative doses, which could also be a potential complication of hypofractionated re-irradiation. In our study, all patients were treated with re-irradiation using doses of 1.8–2.5 Gy per fraction, and no cases of radiation necrosis, intratumoral hemorrhage, or other complications were observed.

In patients with progressive DIPG, the interval between initial RT and progression is a significant prognostic factor, with a longer progression-free interval being associated with a more favorable prognosis [20]. Due to the retrospective nature of much of the existing literature, re-irradiation may be more frequently considered for patients with a longer progression-free interval, potentially introducing selection bias. A key observation in our study is the similarity of median progression-free intervals following the initial RT between patients who received re-irradiation and those who did not (11 months vs. 9 months; p = 0.25). This observation implies that the potential selection bias associated with progression-free intervals has been mitigated in our cohort, thereby enabling a more accurate assessment of the impact of re-irradiation on oncological outcomes.

While our study provides evidence of an OS and clinical benefit associated with re-irradiation in patients who progressed within a similar timeframe after initial RT to those who did not undergo re-irradiation, several limitations must be considered. The retrospective design introduces potential biases, particularly in terms of patient selection and data collection, and the relatively small sample size may limit the generalizability and statistical power of our findings. Additionally, we could not assess the impact of re-irradiation on steroid dependency or quality of life and lacked data on concurrent systemic therapy during re-irradiation. Another limitation was the absence of pre-re-irradiation performance scores. However, in our practice, we also administer re-irradiation to patients with low performance scores, considering that tumor progression may have caused the decline. Therefore, despite the missing data, it is unlikely that only patients with high performance scores were selected for re-irradiation. Despite these challenges, considering the limited literature on this rare and devastating disease, we believe that every contribution to this field advances patient management and provides valuable insights for future research.