Ninety-three patients met inclusion criteria and were evaluated in this retrospective review, with clinical and demographic characteristics presented in Table 1. The mean age of patients in the pre-frail group was 64.0 ± 12.3 with 57 (72%) being females and 22 (28%) being males. The mean age of patients in the frail group was 67.1 ± 10.2 with 10 (71%) being females and 4 (29%) being males. The median radiographic follow up was 15 months (IQR: 9–27) in the pre-frail group compared to 11 months (IQR: 7–20) in the frail group (p = 0.30). The median clinical follow up was 16 months (IQR: 10–27) in the pre-frail group compared to 13 months (IQR: 10–24) in the frail group (p = 0.61). Most patients were classified as prefrail (79, 84.9%) and the remainder classified as a group of frail/severely frail (14, 15.1%). No significant difference was observed between the two groups regarding age, sex, BMI or race.

As expected, patients in the frail cohort had increased rates of hypertension, diabetes, heart failure and the need for assistance with daily living (Table 1). The rates of COPD were similar in both cohorts.

Most patients in both groups had undergone surgical excision prior to SRS with a notably higher proportion in the frail/severely frail group (10, 74.1%) compared to the pre-frail group (43, 54.4%), although this was not found to be significant. The history+ of prior external beam whole brain radiation was less common but still more frequent in the frail/severely frail group (4, 28.6%) versus the pre-frail group (10, 12.7%), although this was also not found to be significant.

Within this cohort of patients, pathology was available for 52 patients (55.9%) and pathology revealed 26 patients with WHO grade I meningiomas and 26 with WHO grade II meningiomas (Table 2). Pre-frail patients had a smaller proportion of confirmed higher grade meningiomas with 23 patients (53.5%) having WHO grade I meningiomas and 20 patients (46.5%) having WHO grade II meningiomas. The frail group demonstrated a predilection towards higher grade meningiomas with 66.7% (n = 6) confirmed as WHO grade II and 33.3% (n = 3) confirmed WHO grade I, although this was not significant. The frail group also demonstrated a significantly (p = 0.021) greater tumor volume (12.1 ± 13.5) vs. the pre-frail group (6.3 ± 7.3). Radiosurgical variables were similar across both cohorts (Table 2).

Kaplan–Meier analysis was then performed to examine probabilities of local, distant and total control. The 1-,2-, and 3-year probability of local control in the pre-frail group was 98.6%, 80.2% and 74.4% respectively. The 1-,2-, and 3-year probability of local control in the frail group was 73.9%, 73.9%, and 73.9% respectively (Fig. 1a). This demonstrated a significant difference between the pre-frail and frail groups with respect to local control (HR 0.22, 95% CI 0.036–1.42, p = 0.01). The 1-, 2-, and 3-year DC probability for pre-frail patients was 98.5%, 94.9%, and 94.9% respectively, while the 1-,2-, and 3-year DC probability for frail patients was 91.6%, 91.6%, and 91.6% respectively (Fig. 1b). This relationship demonstrated no significant difference between the pre-frail and frail group with respect to distant control (HR 0.26, 95% CI 0.01–5.30, p = 0.26). The 1-,2-, and 3-year probability of total control in the pre-frail group was 98.6%, 80.2%, and 74.4% respectively. The 1-, 2-, and 3-year probability of total control was 73.9%, 73.9%, and 73.9% respectively (Fig. 1c).

a Kaplan–Meier plot of local control for all patients at 36 months. b Kaplan–Meier plot of distant control at 36 months. c Kaplan–Meier plot of total control 36 months stratified by frailty status

Kaplan–Meier analysis also demonstrated a significant association (p = 0.02) between frailty and PFS (Fig. 2a). The 1-,2-, and 3-year PFS probability for pre-frail patients was 95.2%, 75.4%, 75.4% respectively. The 1-,2-, and 3-year PFS probability for frail patients was 68.7%, 55.0%, and 36.6% respectively. There was a trend towards diminished OS with increasing frailty after SRS for meningiomas although this did not achieve statistical significance (p = 0.05). The 1-, 2-, and 3-year OS probability for pre-frail patients was 95.5%, 90.5%, and 90.5% respectively. The 1-, 2-, and 3-year OS probability in the frail group was 92.3%, 73.8%, and 49.2% respectively (Fig. 2b).

a Kaplan–Meier plot of progression free survival at 36 months stratified by frailty status. b Kaplan–Meier plot of overall survival at 36 months stratified by frailty status

On univariate analysis (Table 3) there was an association between frailty and progression free survival (PFS) (HR: 2.95 (1.09–7.59); p = 0.033). Prior surgical excision, interval between surgery and radiosurgery and cumulative tumor volume were all associated with earlier tumor progression. In a bivariable model adjusting for interval to radiosurgery, frailty maintained an association with PFS (HR for frailty: 3.45 (1.07–11.07); p = 0.038). Of note, prior surgical excision may represent a clinically significant risk factor of progression.