Abstract
Background
Glioblastoma (GBM) poses therapeutic challenges due to its aggressive nature, particularly for patients with poor functional status and/or advanced disease. Hypofractionated radiotherapy (RT) regimens have demonstrated comparable disease outcomes for this population while allowing treatment to be completed more quickly. Here, we report our institutional outcomes of patients treated with 2 hypofractionated RT regimens: 40 Gy/15fx (3w-RT) and 50 Gy/20fx (4w-RT).
Methods
A single-institution retrospective analysis was conducted of 127 GBM patients who underwent 3w-RT or 4w-RT. Patient characteristics, treatment regimens, and outcomes were analyzed. Univariate and multivariable Cox regression models were used to estimate progression-free survival (PFS) and overall survival (OS). The impact of chemotherapy and RT schedule was explored through subgroup analyses.
Results
Median OS for the entire cohort was 7.7 months. There were no significant differences in PFS or OS between 3w-RT and 4w-RT groups overall. Receipt and timing of temozolomide (TMZ) emerged as the variable most strongly associated with survival, with patients receiving adjuvant-only or concurrent and adjuvant TMZ having significantly improved PFS and OS (P < .001). In a subgroup analysis of patients that did not receive TMZ, patients in the 4w-RT group demonstrated a trend toward improved OS as compared to the 3w-RT group (P = .12).
Conclusions
This study demonstrates comparable survival outcomes between 3w-RT and 4w-RT regimens in GBM patients. Receipt and timing of TMZ were strongly associated with survival outcomes. The potential benefit of dose-escalated hypofractionation for patients not receiving chemotherapy warrants further investigation and emphasizes the importance of personalized treatment approaches.
Keywords: dose escalation, elderly, frail, glioblastoma multiforme, hypofractionated radiotherapy
High-grade gliomas, including glioblastoma (GBM), represent the most common primary malignancies of the central nervous system in adults, comprising approximately 70% of new brain cancer diagnoses.1 The typical management strategy for newly diagnosed GBM (nGBM) is maximal safe resection followed by adjuvant treatment with radiotherapy (RT) and/or temozolomide (TMZ) chemotherapy. The current standard of care for patients below the age of 70 includes maximal safe resection, adjuvant radiation to a dose of 60 Gray (Gy) in 30 fractions with concurrent TMZ, followed by adjuvant TMZ and consideration of maintenance tumor-treating fields. This regimen resulted in a median overall survival (mOS) of 20.9 months.2 However, despite incremental improvements in the treatment paradigm over the past several decades, the prognosis remains poor with mOS ranging from 13.4 to 32.1 months in recent nGBM phase III randomized-controlled trials, with the wide range due to differing patient inclusion criteria. Outcomes for elderly and/or frail patients are especially poor, with an mOS of less than 1 year.2–7
Given these suboptimal survival outcomes, efforts have been made to personalize treatment approaches in nGBM with the goal of minimizing toxicity and time spent in healthcare settings, especially for patients with worse prognoses. A common strategy to minimize treatment time for elderly patients is hypofractionation of adjuvant RT, for which multiple treatment regimens have been evaluated in randomized trials.8 Roa et al. conducted a randomized trial comparing adjuvant conventionally fractionated RT alone (60 Gy in 30 fractions) with a 3-week regimen (40 Gy in 15 fractions) for individuals 60 years and older, and found no significant difference in OS with RT alone using either course.9 Perry et al. compared hypofractionated RT (40 Gy in 15 fractions) alone versus hypofractionated RT with concurrent and adjuvant TMZ in patients aged 65 or older and observed an mOS benefit with the addition of chemotherapy (9.3 months vs 7.6 months, P < .001).10 These results subsequently established hypofractionated RT to 40 Gy in 15 fractions with concurrent and adjuvant TMZ as a standard-of-care treatment option for elderly patients with nGBM.
One potential criticism of the RT regimen used by Perry et al. is a lower biologically effective dose (BED) compared to 60 Gy in 30 fractions used in Stupp et al. (50.67 Gy vs 72 Gy, respectively, using an alpha/beta of 10). Therefore, some groups have investigated dose-escalated hypofractionated regimens such as Perlow et al., who recently described outcomes after 52.5 Gy in 15 fractions (BED 70.9 Gy).11 In a cohort of 62 elderly patients with GBM, the authors observed encouraging OS compared to historical controls (mOS 10.3 months).12 The aim of this investigation is to describe our large institutional experience with 2 hypofractionated RT regimens (40 Gy in 15 fractions and a dose-escalated strategy of 50 Gy in 20 fractions) for patients with nGBM. Our results may improve personalization of treatment recommendations for this vulnerable patient population.
Methods
Patient Selection and Evaluation
Following MD Anderson Cancer Center Institutional Review Board approval, we retrospectively reviewed electronic medical records of individuals receiving hypofractionated RT for the treatment nGBM at our institution from 2012 to 2022. For the purposes of this study, hypofractionated RT regimens included a prescription dose of either 50 Gy delivered in 20 fractions (referred to hereafter as 4w-RT) or 40 Gy in 15 fractions (referred to hereafter as 3w-RT). For eligible patients, we collected demographic-, disease-, and treatment-related information including: age while receiving RT, Karnofsky performance status (KPS) score13 at initiation of RT, Charlson comorbidity index,14 sex, ethnicity, tumor laterality (categorized as unilateral vs bilateral), type of surgery (gross total resection [GTR], subtotal resection [STR], or biopsy alone), RT regimen, isocitrate dehydrogenase (IDH) mutation status, O6-methylguanine-DNA-methyltransferase promoter (pMGMT) methylation status, TMZ exposure (none, concurrent with RT only, adjuvant only, or concurrent with RT + adjuvant), progression-free survival (PFS) period, and OS. In this study, nGBM collectively refers to patients with WHO grade 4 glioma, regardless of IDH mutation status (pre-WHO 2021 reclassification),15 which permitted comparison of study results to those from recently published trials with similar patient cohorts.
Treatment Details
All patients underwent either biopsy alone or maximal safe resection, which could include STR or GTR. One patient underwent laser interstitial thermal therapy with biopsy alone as primary surgical management. Adjuvant RT was delivered in daily fractions to achieve the prescribed dose using a thermoplastic mask for daily immobilization of the head. Treatment targets were delineated with the assistance of diagnostic MRI fused to the planning CTs in the treatment planning system.
Targets included the gross tumor volume (GTV), which encompassed the postoperative tumor bed and any residual contrast-enhancing disease. For individuals receiving 3w-RT, a clinical target volume (CTV) was defined by a 1.5-cm uniform expansion from the GTV, respecting anatomic barriers of microscopic disease spread including bone and dural reflections. Additional expansion or intentional inclusion of fluid-attenuated inversion recovery (FLAIR) was at the discretion of the treating physician. For 3w-RT, the CTV was prescribed as either a uniform 40.05 Gy or a lower dose (minimum of 30 Gy) with a simultaneous integrated boost (SIB) to 40.05 Gy to the GTV. For those receiving 4w-RT, the CTV was defined as a 2-cm anatomically confined expansion from GTV, which could also encapsulate additional FLAIR signal at the discretion of the treating physician. Typically, patients receiving 4w-RT were prescribed 40 Gy to the CTV with an SIB to 50 Gy to the GTV; however, a minority of patients were treated to higher CTV doses or a uniform dose of 50 Gy to the entire CTV at the discretion of the treating physician. Planning target volumes were created using a 3-mm uniform expansion from GTV and/or CTV.
Systemic therapy could include TMZ given concurrently with RT, in the adjuvant setting after completion of RT, or both. Exposure to other therapies including anti-angiogenic agents or other cytotoxic therapies was considered an exclusion criterion for the purposes of this study.
Outcomes
The primary outcomes were PFS and OS, defined as time in months from completion of RT to radiographic disease progression or death, respectively. Disease progression was designated according to radiologist reports from posttreatment MRI scans and was categorized as either local failure within the RT field or distant failure elsewhere. PFS was calculated as the time to any (local or distant) disease progression or death.
Statistical Analysis
Descriptive statistics were used to characterize the identified patient population. Continuous variables are presented as median (interquartile range [IQR]). Nonparametric testing was used to compare differences across populations. Chi-square testing was used to compare proportions of counts between groups. Cox proportional hazard modeling was used to evaluate differences in time-to-event outcomes including progression and death. Univariate comparisons were performed initially and factors with significant association were selected to develop a multivariable model inclusive of the variable of interest (RT fractionation scheme) to identify associations with PFS and/or OS. Hazard ratios (HRs) were estimated and a forest plot was generated to visualize the multivariable model. The Kaplan–Meier method was used to describe differences in survival outcomes and the log-rank test was used to evaluate differences in Kaplan–Meier survival curves. Patients who did not experience a progression or death event were censored at last follow-up time for Kaplan–Meier survival analysis. Statistical significance was defined as P < .05. All statistical analyses were performed with R 4.3.0 (R Foundation for Statistical Computing, Vienna, Austria).
Results
A total of 127 patients were identified and included in the analysis. Baseline demographic and treatment characteristics are summarized in Table 1. Median age was 69.9 years (IQR, 61.5–75.5) and most patients were male (57%). Seventy-three patients (57%) received 4w-RT and 54 (43%) received 3w-RT. Compared to patients receiving 4w-RT, individuals receiving 3w-RT were older (median 66.7 vs 71.7 years, respectively, P = .012) and more likely to have higher Charlson comorbidity index (median 4 vs 5, P = .007). There were also less likely to receive concurrent-only TMZ (21% vs 7.4%), less likely to receive concurrent and adjuvant TMZ (41% vs 24%), and more likely to not receive any TMZ (22% vs 48%, P = .005). There were no significant differences in sex, IDH mutation status, MGMT methylation status, KPS, extent of resection, or tumor bilaterality between the groups.
Table 1.
Baseline Characteristics of the Study Population, With Results Stratified by Radiotherapy Regimen
| Characteristic | Overall | % | 4w-RT | % | 3w-RT | % | P Value |
|---|---|---|---|---|---|---|---|
| n = 127 | n = 73 | n = 54 | |||||
| Age, years (median [Q1–Q3]) | 69.9 [61.5–75.5] | 66.7 [59.9–74.1] | 71.7 [67.3–75.9] | .012 | |||
| Sex | .86 | ||||||
| Male | 72 | 57 | 42 | 58 | 30 | 56 | |
| Female | 55 | 43 | 31 | 42 | 24 | 44 | |
| IDH1/2 mutation status | .99 | ||||||
| Wild type | 79 | 62 | 45 | 62 | 34 | 63 | |
| Unknown | 48 | 38 | 28 | 38 | 20 | 37 | |
| pMGMT methylation status | 0 | .88 | |||||
| Negative | 21 | 17 | 9 | 12 | 12 | 22 | |
| Positive | 20 | 16 | 8 | 11 | 12 | 22 | |
| Unknown | 86 | 68 | 56 | 77 | 30 | 56 | |
| Karnofsky performance status score | .1 | ||||||
| ≥70 | 87 | 69 | 54 | 74 | 33 | 61 | |
| <70 | 40 | 31 | 19 | 26 | 21 | 39 | |
| Charlson comorbidity index (median [Q1–Q3]) | 5 [4–6] | 4 [4–6] | 5 [5–6] | .007 | |||
| Temozolomide | .005 | ||||||
| Concurrent only | 19 | 15 | 15 | 21 | 4 | 7.4 | |
| Adjuvant only | 23 | 18 | 12 | 16 | 11 | 20 | |
| Concurrent and adjuvant | 43 | 34 | 30 | 41 | 13 | 24 | |
| None | 42 | 33 | 16 | 22 | 26 | 48 | |
| Extent of resection | .17 | ||||||
| Biopsy | 40 | 31 | 19 | 26 | 21 | 39 | |
| Subtotal resection | 54 | 43 | 36 | 49 | 18 | 33 | |
| Gross total resection | 33 | 26 | 18 | 25 | 15 | 28 | |
| Bilateral tumor | .37 | ||||||
| Yes | 63 | 50 | 39 | 53 | 24 | 44 | |
| No | 64 | 50 | 34 | 47 | 30 | 56 |
Abbreviations: IDH, isocitrate dehydrogenase; pMGMT methylation status, pathologic O6-methylguanine-DNA methyltransferase promoter methylation status; 3w-RT, 40 Gy in 15 fractions delivered over 3 weeks; 4w-RT, 50 Gy in 20 fractions delivered over 4 weeks.
The median PFS and OS for the study population were 5.5 months and 7.7 months, respectively. Table 2 summarizes univariate and multivariable Cox regression models to predict PFS and OS. In both univariate and multivariable comparisons, there was no significant difference in PFS or OS between the 3w-RT and 4w-RT groups. The only variable independently associated with PFS or OS was receipt and timing of TMZ. Compared to individuals receiving no TMZ, those receiving adjuvant TMZ only (HR 0.43 [0.25–0.73], P = .0017) and those receiving concurrent and adjuvant TMZ (HR 0.33 [0.21–0.53], P < .001) had greater PFS. Despite univariate associations with KPS, TMZ receipt/timing, extent of resection, and tumor bilaterality, only TMZ receipt/timing persisted as significant predictors of OS in a multivariable model (Figure 1). Compared to individuals receiving no TMZ, those receiving adjuvant TMZ only (HR 0.44 [0.25–0.77], P = .004) and those receiving concurrent and adjuvant TMZ (HR 0.32 [0.2–0.51], P < .001) had greater OS. Survival outcomes for the full study population with stratification by RT and chemotherapy regimen are displayed in Figure 2.
Table 2.
Univariate and Multivariable Cox Regression Models for Progression-Free and Overall Survival
| Endpoint: Progression-Free Survival | Endpoint: Overall Survival | |||||||
|---|---|---|---|---|---|---|---|---|
| Univariate Cox | Multivariable Cox | Univariate Cox | Multivariable Cox | |||||
| Characteristic | HR (95% CI) | P Value | HR (95% CI) | P Value | HR (95% CI) | P Value | HR (95% CI) | P Value |
| Radiotherapy regimen | ||||||||
| 3w-RT | Ref | Ref | Ref | Ref | ||||
| 4w-RT | 1.16 (0.67–2.0) | .6 | 0.73 (0.49–1.08) | .12 | 0.78 (0.54–1.13) | .2 | 0.74 (1.33–1.13) | .17 |
| Age (continuous) | 0.98 (0.92–1.01) | .2 | 1 (0.98–1.02) | .99 | ||||
| Sex | ||||||||
| Female | Ref | Ref | ||||||
| Male | 1.26 (0.75–2.13) | .4 | 1.18 (0.82–1.7) | .4 | ||||
| IDH1/2 mutation status | ||||||||
| WT | Ref | Ref | ||||||
| Unknown | 1.35 (0.8–2.23) | .3 | 1.25 (0.86–1.81) | .2 | ||||
| pMGMT methylation status | ||||||||
| No | Ref | Ref | ||||||
| Yes | 0.69 (0.26–1.8) | .4 | 0.73 (0.38–1.39) | .3 | ||||
| Unknown | 0.84 (0.57–2.47) | .6 | 0.89 (0.54–1.46) | .6 | ||||
| Karnofsky performance status score | ||||||||
| ≥70 | Ref | Ref | Ref | |||||
| <70 | 1.58 (0.88–2.82) | .1 | 1.64 (1.1–2.43) | .01 | 0.82 (0.53–1.26) | .37 | ||
| Charlson comorbidity index (continuous) | 0.85 (0.71–1.02) | .084 | 0.98 (0.87–1.11) | .8 | ||||
| Temozolomide | ||||||||
| None | Ref | Ref | Ref | Ref | ||||
| Concurrent only | 2.8 (1.13–6.9) | .027 | 1.34 (0.74–2.43) | .33 | 1.16 (0.66–2) | .6 | 1.29 (0.7–2.39) | .42 |
| Adjuvant only | 0.69 (0.32–1.51) | .4 | 0.43 (0.25–0.73) | .0017 | 0.42 (0.25–0.71) | .001 | 0.44 (0.25–0.77) | .004 |
| Concurrent and adjuvant | 0.6 (0.3–1.19) | .1 | 0.33 (0.21–0.53) | <.001 | 0.32 (0.2–0.51) | <.001 | 0.34 (0.2–0.58) | <.001 |
| Extent of resection | ||||||||
| Biopsy | Ref | Ref | Ref | |||||
| Subtotal resection | 1.37 (0.72–2.59) | .3 | 0.83 (0.55–1.27) | .4 | 0.69 (0.43–1.09) | .11 | ||
| Gross total resection | 0.73 (0.36–1.47) | .4 | 0.5 (0.31–0.81) | .004 | 0.62 (0.36–1.09) | .1 | ||
| Bilateral tumor | ||||||||
| No | Ref | Ref | Ref | |||||
| Yes | 1.39 (0.83–2.32) | .2 | 1.7 (1.18–2.45) | .004 | 1.28 (0.82–2) | .27 | ||
Abbreviations: CI, confidence interval; HR, hazard ratio; IDH, isocitrate dehydrogenase; pMGMT methylation status, pathologic O6-methylguanine-DNA methyltransferase promoter methylation status; WT, wild type; 3w-RT, 40 Gy in 15 fractions delivered over 3 weeks; 4w-RT, 50 Gy in 20 fractions delivered over 4 weeks.
Figure 1.
Forest plot displaying estimated hazard ratios for death for covariates included in the final multivariable model. Abbreviations: 3w-RT, 40 Gy in 15 fractions delivered over 3 weeks; 4w-RT, 50 Gy in 20 fractions delivered over 4 weeks; GTR, gross total resection; KPS70, Karnofsky performance status of 70 or greater; STR, subtotal resection.
Figure 2.
Kaplan–Meier estimates of overall survival for the full study population (A), with results stratified by radiotherapy regimen (B) and receipt and schedule of TMZ (C). Abbreviation: TMZ, temozolomide.
Given the poorer survival in patients not receiving TMZ, we performed an exploratory analysis evaluating the impact of RT schedule in this population. Kaplan–Meier survival curves for individuals not receiving TMZ with stratification by RT regimen are shown in Figure 3B. Although not reaching statistical significance (P = .12), there was numerically improved OS in those patients receiving 4w-RT in this subpopulation (mOS 3.2 months vs 5.9 months for 3w-RT and 4w-RT, respectively). Further, when evaluating specifically the population of patients that did not receive adjuvant TMZ (but may have received concurrent TMZ), individuals undergoing 4w-RT experienced improved OS (P = .06; Figure 3C).
Figure 3.
Kaplan–Meier estimates of overall survival for individuals receiving any TMZ (A), those not receiving any TMZ (B), and those not receiving adjuvant TMZ (C), with results stratified by radiotherapy regimen. Abbreviation: TMZ, temozolomide.
Discussion
In this single-institution retrospective analysis, we compared outcomes for individuals with nGBM receiving 2 hypofractionated RT regimens (3w-RT or 4w-RT) and found no significant differences in oncologic outcomes between groups. The mOS was 7.7 months, consistent with results of many studies of hypofractionated RT for frail and elderly patients with GBM. After accounting for potential confounders including performance status, age, and extent of surgical resection, the only significant predictor of OS in this cohort was receipt and timing of TMZ. Patients who received adjuvant or concurrent and adjuvant TMZ experienced significantly improved PFS and OS. In the subpopulation of patients who did not receive any TMZ, there was numerically improved OS for those patients receiving 4w-RT, although differences were not statistically significant. Although these results may partially reflect unaccounted treatment selection bias, in general, they suggest similar outcomes for individuals receiving either regimen of hypofractionated RT. Personalization of treatment may be possible according to other factors such as planned chemotherapy, social factors, and performance status.
Many of the clinical trials that have established the current standard of care in nGBM, such as the Stupp trial, excluded patients over the age of 65 years or those with poor performance status and, therefore, generalization of trial findings to these patients is difficult. With many estimates of survival of less than 1 year among this vulnerable population, it is critical that clinicians optimally balance treatment benefits and harms to maximize patient quality and quantity of life. Conventionally fractionated RT approaches, such as the standard 60 Gy in 2 Gy fractions, require 6 weeks of therapy that may compromise a patient’s ability to optimally spend their time with family and loved ones. Furthermore, these regimens may lead to higher levels of toxicity that may be particularly debilitating for vulnerable populations. However, omission of RT entirely has been proven to reduce OS in elderly patients without improving quality of life compared to supportive care alone.16 Hypofractionation therefore represents a valuable tool to optimize the therapeutic ratio in these clinical situations through a reduction in time spent in healthcare environments and potential decrease in toxicity burden.17
One common strategy for hypofractionation of adjuvant RT is the regimen adopted from Perry et al., which prescribed a dose of 40.05 Gy in 15 fractions, thus completing in 3 weeks (referred to here as 3w-RT). Given the lower BED10 of this regimen compared to the RT regimen used in the Stupp protocol (50.67 Gy vs 72 Gy), there has recently been growing interest in dose-escalated hypofractionation to achieve BED comparable to standard conventionally fractionated approaches. Perlow et al. showed encouraging preliminary outcomes from a pooled analysis of 62 elderly patients with nGBM receiving 52.5 Gy in 15 fractions (BED 70.9) on early-phase clinical trials, with an mOS of 10.3 months. The authors concluded that dose-escalated hypofractionation to similar BED of conventional approaches yields good outcomes overall compared to historical studies.12
At our institution, we have utilized multiple approaches for hypofractionation including the 3w-RT regimen as well as a more aggressive approach delivering 50 Gy in 20 fractions (4w-RT), which has a BED10 (62.5 Gy) closer to that of the RT regimen using in the Stupp trials. The 4w-RT regimen was introduced in the late 1980s as an option for elderly patients or those with poor performance status at a time when there was no consensus for an optimal hypofractionation strategy. Our experience with 4w-RT yielded encouraging oncologic outcomes without apparent increased toxicity and thus became a standard option for select patients at our institution.18
Although the 4w-RT regimen does not have equivalent BED10 compared to the 52.5 Gy regimen described in the analysis by Perlow et al. (70.9 Gy), it is significantly higher equivalent dose compared to the popular Perry regimen. Therefore, particularly given our relatively large sample size compared to other similar investigations, this study could provide hypothesis-generating data regarding the impact of dose escalation in the setting of hypofractionated RT for nGBM. Overall, we found no significant differences in oncologic outcomes between individuals receiving either 3w-RT or 4w-RT, even when controlling for differences in baseline characteristics of the populations such as performance status, tumor bilaterality, extent of resection, and receipt of chemotherapy. These results are consistent with a large body of literature demonstrating no benefit with dose-escalated RT for nGBM, particularly in the era of modern chemotherapy.12–14
Although RT regimen did not appear to be associated with oncologic outcomes in this series overall, there was a strong association with receipt and timing of TMZ. Individuals who received either adjuvant TMZ, or both concurrent and adjuvant TMZ, experienced significantly improved PFS and OS (Figures 1 and 2C). One potential explanation for this finding is confounding wherein individuals with better-expected prognosis overall (eg, related to performance status, age, comorbidities, etc.) would be more likely to tolerate and receive chemotherapy. In that situation, any survival differences would be difficult to definitively attribute to treatment differences alone. We did attempt to address this potential confounding through the creation of a multivariable model, but any model may insufficiently capture real-world covariates impacting endpoints of interest. Additionally, in this study, concurrent TMZ with RT but without adjuvant TMZ was not associated with improved outcomes. This may represent a selection bias of patients who were planned for concurrent and adjuvant TMZ, but did poorly during RT or developed TMZ toxicity, precluding adjuvant TMZ. Nonetheless, given the limitations of retrospective data, our experience does suggest that chemotherapy has a significant impact on outcomes for these patients. These results additionally corroborate the findings of the Perry trial, which demonstrated a benefit of TMZ in the setting of hypofractionated RT.10
Given the strong association between chemotherapy receipt and survival, we performed a secondary analysis to explore the impact of various RT regimens among populations with especially poor prognosis, namely those that did not receive TMZ. Despite the low absolute number of these patients, there did appear to be a trend toward improved survival for individuals receiving the higher-RT dose 4w-RT regimen. While these results do not demonstrate definitive evidence for a clinically meaningful benefit, they do provide early hypothesis-generating data to suggest that personalization of RT in this situation is not only feasible but potentially beneficial. Furthermore, these findings corroborate a recent systematic review and meta-analysis of 22 studies representing 2198 patients with GBM also suggesting that dose-escalated RT in the absence of chemotherapy may improve outcomes, regardless of fraction size.19 Multidisciplinary discussion between medical oncologists and radiation oncologists is critical to optimally integrate multimodality therapy and optimize the therapeutic ratio for these patients. We recommend consideration of dose-escalated hypofractionated RT specifically for patients with GBM who might be expected to have limited tolerance of TMZ (eg, those with low baseline blood counts, immunosuppression, low body mass index/frailty, etc.).
This study does have important limitations related to experimental design. First, as with any observational nonrandomized study, it is impossible to define causal relationships between treatment regimens and oncologic outcomes. For example, it is very likely that RT regimens were chosen by physicians based on clinical factors related to overall health and ability to tolerate escalated therapy. We did attempt to account for such confounding using a multivariable model including covariates of interest collected in the data set; nonetheless, it is possible that some confounding factors that may impact treatment selection and outcomes were not accounted for in our analysis. Furthermore, there are several approaches that could be utilized for causal inference using observational data. Aside from multivariable modeling, propensity score-matching is a common strategy that could have been utilized to analyze this cohort and compare outcomes for individuals receiving 4w-RT and 3w-RT. There is no consensus regarding an optimal strategy in these situations, and most studies comparing approaches show similar performance.20,21 Here, given a relatively small cohort with limited ability to accurately match large numbers of patients, we elected to utilize a multivariable modeling approach, although other analytic methods could have been equally valid. Second, for some clinical variables of interest, such as pMGMT methylation status and IDH mutation status, there was a percentage of incomplete data in the electronic medical records. Nonetheless, we were unable to observe any significant differences in outcomes with the available data. Lastly, although the sample represents the treatment experience of a large academic center, the overall sample size is somewhat limited for specific subpopulations within the cohort.
Conclusions
In this analysis of 127 patients with GBM receiving hypofractionated RT, we observed no differences in survival outcomes for individuals receiving 3w-RT and 4w-RT. The strongest predictor of improved survival was receipt of adjuvant TMZ, an effect which persisted when accounting for other covariates such as age, comorbidity index, and performance status. There may be a potential benefit of dose-escalated hypofractionated RT particularly for individuals not planned to receive chemotherapy. These results may be used to improve shared decision-making between providers and patients in order to maximize quality and quantity of life.
Contributor Information
Thomas H Beckham, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Michael K Rooney, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Mary F McAleer, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Amol J Ghia, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Martin C Tom, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Subha Perni, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Susan McGovern, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
David Grosshans, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Caroline Chung, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Chenyang Wang, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Brain De, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Todd Swanson, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Arnold Paulino, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Wen Jiang, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Sherise Ferguson, Department of Neurosurgery, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Chirag B Patel, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Jing Li, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Debra N Yeboa, Department of Radiation Oncology, CNS/Pediatrics Section, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Funding
C.B.P. is a McNair Scholar supported by the McNair Medical Institute at The Robert and Janice McNair Foundation.
Conflict of Interest
None declared.
References
- 1. Ostrom QT, Cioffi G, Waite K, Kruchko C, Barnholtz-Sloan JS.. CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2014–2018. Neuro Oncol. 2021;23(12 suppl 2):iii1–iii105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Stupp R, Taillibert S, Kanner A, et al. Effect of tumor-treating fields plus maintenance temozolomide vs maintenance temozolomide alone on survival in patients with glioblastoma: a randomized clinical trial. JAMA. 2017;318(23):2306–2316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Nunna RS, Khalid SI, Patel S, et al. Outcomes and patterns of care in elderly patients with glioblastoma multiforme. World Neurosurg. 2021;149:e1026–e1037. [DOI] [PubMed] [Google Scholar]
- 4. Stupp R, Taillibert S, Kanner AA, et al. Maintenance therapy with tumor-treating fields plus temozolomide vs temozolomide alone for glioblastoma: a randomized clinical trial. JAMA. 2015;314(23):2535–2543. [DOI] [PubMed] [Google Scholar]
- 5. Lassman AB, Pugh SL, Wang TJC, et al. Depatuxizumab mafodotin in EGFR-amplified newly diagnosed glioblastoma: a phase III randomized clinical trial. Neuro Oncol. 2023;25(2):339–350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Omuro A, Brandes AA, Carpentier AF, et al. Radiotherapy combined with nivolumab or temozolomide for newly diagnosed glioblastoma with unmethylated MGMT promoter: an international randomized phase III trial. Neuro Oncol. 2023;25(1):123–134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Liau LM, Ashkan K, Brem S, et al. Association of autologous tumor lysate-loaded dendritic cell vaccination with extension of survival among patients with newly diagnosed and recurrent glioblastoma: a phase 3 prospective externally controlled cohort trial. JAMA Oncol. 2023;9(1):112–121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. de Melo SM, Marta GN, Latorraca C de OC, et al. Hypofractionated radiotherapy for newly diagnosed elderly glioblastoma patients: a systematic review and network meta-analysis. PLoS One. 2021;16(11):e0257384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Roa W, Brasher PMA, Bauman G, et al. Abbreviated course of radiation therapy in older patients with glioblastoma multiforme: a prospective randomized clinical trial. J Clin Oncol. 2004;22(9):1583–1588. [DOI] [PubMed] [Google Scholar]
- 10. Perry JR, Laperriere N, O’Callaghan CJ, et al. Short-course radiation plus temozolomide in elderly patients with glioblastoma. N Engl J Med. 2017;376(11):1027–1037. [DOI] [PubMed] [Google Scholar]
- 11. Perlow HK, Yaney A, Yang M, et al. Dose-escalated accelerated hypofractionation for elderly or frail patients with a newly diagnosed glioblastoma. J Neurooncol. 2022;156(2):399–406. [DOI] [PubMed] [Google Scholar]
- 12. Perlow HK, Prasad RN, Yang M, et al. Accelerated hypofractionated radiation for elderly or frail patients with a newly diagnosed glioblastoma: a pooled analysis of patient-level data from 4 prospective trials. Cancer. 2022;128(12):2367–2374. [DOI] [PubMed] [Google Scholar]
- 13. Karnofsky D, Burchenal J.. The evaluation of chemotherapeutic agents against neoplastic disease. Cancer Res. 1948;8(8):388–389. [Google Scholar]
- 14. Charlson ME, Pompei P, Ales KL, MacKenzie CR.. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–383. [DOI] [PubMed] [Google Scholar]
- 15. Organisation mondiale de la santé, Centre international de recherche sur le cancer, eds. Central Nervous System Tumours. 5th ed. Lyon, France: International Agency for Research on Cancer; 2021. [Google Scholar]
- 16. Keime-Guibert F, Chinot O, Taillandier L, et al. Radiotherapy for glioblastoma in the elderly. N Engl J Med. 2007;356(15):1527–1535. [DOI] [PubMed] [Google Scholar]
- 17. Clarke JW, Chang EL, Levin VA, et al. Optimizing radiotherapy schedules for elderly glioblastoma multiforme patients. Expert Rev Anticancer Ther. 2008;8(5):733–741. [DOI] [PubMed] [Google Scholar]
- 18. Chang EL, Yi W, Allen PK, et al. Hypofractionated radiotherapy for elderly or younger low-performance status glioblastoma patients: outcome and prognostic factors. Int J Radiat Oncol Biol Phys. 2003;56(2):519–528. [DOI] [PubMed] [Google Scholar]
- 19. Singh R, Lehrer EJ, Wang M, et al. Dose-escalated radiation therapy for glioblastoma multiforme: an international systematic review and meta-analysis of 22 prospective trials. Int J Radiat Oncol Biol Phys. 2021;111(2):371–384. [DOI] [PubMed] [Google Scholar]
- 20. Biondi-Zoccai G, Romagnoli E, Agostoni P, et al. Are propensity scores really superior to standard multivariable analysis? Contemp Clin Trials. 2011;32(5):731–740. [DOI] [PubMed] [Google Scholar]
- 21. Cioci AC, Cioci AL, Mantero AMA, et al. Advanced statistics: multiple logistic regression, cox proportional hazards, and propensity scores. Surg Infect. 2021;22(6):604–610. [DOI] [PubMed] [Google Scholar]