Does age really matter? Radiotherapy in elderly patients with glioblastoma, the Munich experience

With a median age of 64 years, glioblastoma (GBM) patients are in nearly half of cases at an age, that frequently defines them as “aged” or “elderly” [1, 2]. Generally, the present standard of care for GBM patients includes involved field radiotherapy (RT) as well as concomitant as well as 6 cycles of adjuvant chemotherapy with temozolomide (TMZ) and goes back to the study from Stupp et al. in 2005 [3]. The patient cohort, however, was limited to an age of equal or less than 70 years and a post-hoc analysis of this cohort found a negative correlation between the patient’s age and the benefit from a combined regimen [3, 4]. An anticipated increased likelihood of adverse events of TMZ in elderly patients might be one explanation for this finding [5]. Notably, several mono-institutional reports as well as data-base-studies have demonstrated that elderly GBM patients treated with standard-RT plus TMZ do have a longer survival compared to patients treated with alternative or reduced regimens, especially after extensive resection of the tumor [6‐8].

As the positive effect of RT in elderly patients is not a matter of debate anymore [9], several alternative dosing-regimens have been tested in prospective trials. Exemplary, Roa and colleagues demonstrated non-inferiority of a 3 week regimen compared to the 6 week standard regimen of radiotherapy [10]. Recently, Roa and colleagues demonstrated, that also an even more hypofractionated one-week-regimen is equal-effective in elderly and frail patients compared to the formerly described mildly hypofractionated 3 week course [11]. Of note, neither of these two regimens were tested together with concomitant chemotherapy.

TMZ, on the other hand, was tested to be non-inferior to RT in elderly patients with a methylated MGMT-promotor in the NOA-08-Trial: The efficacy of RT did not depend on the MGMT-status of the treated high grade gliomas. Also this trial did not test radiochemotherapy in elderly patients [2]. This gap recently was closed by prospective data from an international phase III trial. By comparing hypofractionated RT with concomitant TMZ followed by up to 6 adjuvant cycles of TMZ, the authors demonstrated a significant advantage accompanied by a tolerable toxicity profile also for elderly patients treated with the combined regimen, independently from age and using a very inclusive paradigm [12].

While the current evidence strongly supports the role of loco-regional treatments in elderly patients, too, population based studies demonstrate a positive correlation between age and the treatment by best supportive care only, hinting to a possible under-treatment of elderly patients [7]. Hence, there must be a difference between aged frail, almost palliative patients and extremely fit and active elderly, which arguments against age only as a decision making tool.

In the present article, we report on our experience in treating elderly patients with RT and RChT and confirm that a combined modality treatment with radiochemotherapy with TMZ results in a longer survival, independently from the age but dependent from the performance status of the patients.

GBM almost always leads to the loss of independence by increasingly developing disabilities throughout the course of disease; the loss of independence is mostly due to progression of the disease [14]. RT, chemotherapy as well as combined modality treatments have shown to prolong the progression free survival and to increase the OS, too [2, 15, 16]. Based on reports about increased toxicities of either of these treatments in elderly patients, mono-therapeutic regimens have emerged [2, 10, 11]. All of these aim on minimizing the burden by the treatment, and, all of them have demonstrated mOS between 6.4 and 9.6 months in elderly people [2, 10]. Inclusion criteria of these trials were mostly based on age (65 years or older) and a modest to good performance status (Karnofsky Performance Score (KPS) of at least 60% or ECOG of 2 or less).

The median age of the NOA-08 trial, a trial comparing chemotherapy to RT in elderly patients, was 71 vs. 72 years, the median OS was 8.6 for TMZ and 9.4 months for RT, p = 0.033. The trial also analysed the impact of MGMT promotor methylation and described the predictive value for MGMT for the efficacy of TMZ. Similar results were reported from the Nordic Trial, which furthermore reported a small but significant positive effect of either TMZ or hypofractionated RT to a standard fractionated RT [17]. The trial included patients with an age of 60 years or older, the difference described above was more pronounced within the group of patients older than 70 years. Rao et al. randomized patients with a minimum age of 60 years and a mean age of 72.4 and 71 years to receive either a 6 week normofractionated or a 3 week hypofractionated regimen. Both groups had a median KPS of 70%. The trial was closed earlier due to high similarity between the two arms. The trial demonstrated equal efficacy of both dosing schemes, with a median OS of 5.1 and 5.6 months. Notably, Gross Total Resection (GTR) was achieved in only 4.2 and 14.6% of the cases, almost 40% received biopsy only [10]. A mono-institutional report from Ontario, Canada, reported about hypofractionated RT with and without concurrent TMZ. GTR was achieved in one third of the patients. The median survival was superior for the mono-RT group (6.9 vs. 9.3 months), yet the difference was not significant. Similar to our results, also elderly patients had a significant benefit from salvage therapies (5.7 vs. 13.3 months) [18]. Already in 2008, Combs et al. demonstrated the efficacy and safety of RChT for patients older than 65 years. This cohort had a median survival of 11 months, a subgroup with gross total resection had a median survival of 18 months [8]. Only recently, results from the randomized phase III EORTC 26062–22061 / TROG 08.02-Trial were presented by Perry and colleagues. 562 patients were randomized between a hypofractionated RT up to 40.05 Gy alone or in combination with concurrent TMZ. In opposite to the retrospective data from Cao et al., this trial showed a significant advantage for the combined modality treatment (7.6 vs 9.3 months). Importantly, also patients with non-methylated MGMT-promotor benefited from concurrent radiochemotherapy [12]. Unfortunately, the article did not report whether also patients with ECOG 2 did profit from the combined modality treatment. In our cohort, albeit not significant, patients with an ECOG 1 did profit more than patients with ECOG 2. In our view, the radiotherapy treatment regimen could be an important reason for this difference. Hypofractionated radiotherapy was shown to be equally effective in elderly and frail patients not eligible to standard fractionated treatment [10, 11, 17], but concerns about the long term safety have been raised [19]. On the other hand, nether safety nor efficiency of hypofractionated and standard-fractionated RChT have been compared directly. Taken that in hypofractionated treatments the concomitant chemotherapy time it cut in half, medically fit patients (ECOG 0 and 1) treated according to the Perry-Study can be considered to be under risk of being undertreated. A standard fractionated RChT therefore should be deemed to be the standard of care for elderly patients with a good performance status. This concepts has been published previously, stressing the necessity of proactive treatment in medically fit patients above age 65 years [8].

We also analysed the survival of patients after diagnosis for objective progressive disease. Patients which were treated for recurrent GBM, either by surgery, re-irradiation or chemotherapy, did have a longer survival compared to patients receiving best supportive care (BSC) in this situation. Prospective trials comparing BSC and re-irradiation or any other salvage-strategy in elderly patients are scarce; only one article analysed this topic and concluded, that salvage treatment for recurrent GBM could be beneficial [20]. This comparison as well as our analysis are influenced by a selection bias, as patients with a better performance score are more likely to undergo a salvage treatment than patients with a poor OS. Concerning re-irradiation, an interval of less than 6 months between the first course of radiotherapy and progressive disease is usually believed as to short to undergo re-irradiation. As RChT can result in progression free survival of more than 6 months in elderly patients, especially when MGMT promotor hypermethylation is present [12], re-irradiation could become more frequent. Further studies, especially on the safety and efficacy of re-irradiation, are therefore highly recommended.

Our data underline the role of an intensive early treatment as well as of salvage treatments for older patients. Especially patients treated with standard-fractionated RChT as well as adjuvant chemotherapy had a median OS that was similar to younger patients. This is in-line with the conclusion from a SEER-based analysis from 2015 [7]. Reasons for this might be the good physical status of patients within this cohort as well as the high amount of patients with methylated MGMT promotors. As the performance score but not age significantly related to the treatment decision and to the outcome, monotherapy should only be considered for patients older than >70 years and presenting with a lower performance score. This algorithm is also in line with the resent guidelines for the treatment of GBM from the American Society of Clinical Oncology (ASCO) [21]. Notably, when reviewing earlier studies or studies from other geographic areas, the huge geographic differences of life expectancies have to be taken into account. Exemplarily, the life expectancy within the western world approximately increases every 20 years by 5 years, currently reaching an average life expectancy of 80 years, compared to an average life expectancy of 66 years in India.

Combination of RT and chemotherapy in elderly patients, independently of fractionation, has a good efficacy also in elderly patients and should be considered even in higher age but with taking the performance status into account. Therefore, treatment decision should be made based not only on age in order to prevent undertreatment in elderly patients.