PubMed search and Eligibility screening
The Medline and Embase search identified 1024 records. Out of these, 913 citations were discarded following abstract review in view of duplicates, irrelevance to the study objective, nature of publication (review articles, editorials, letter to editor, case reports) and non-availability of original data. The full text of the remaining 111 papers were examined in detail and further 79 publications were eliminated in adherence to the inclusion criteria. The reasons for exclusion for each publication are reported in Supplementary Table 1. Thirty-two studies, thus, were eligible for data extraction and analysis. The 10 studies selected for MS for small VS are detailed in Table 1.[1, 10, 16, 19, 35, 43, 46, 47, 49, 50] The 23 studies selected for the SRS for small VS are detailed in Table 2.[5, 32, 17, 36, 42, 48, 38, 44, 3, 26, 8, 22, 16, 20, 25, 45, 4, 28, 31, 37, 7, 18, 11] One study was a matched cohort comparison and was included in both arms.[16]
Table 1
List of excluded publications with reasons for exclusion
Author, Publication Year | Enrollment Period | Type of study | Patient's analyzed | Age (years) | F/U (months) | Approach | Mean Tumor Size |
| | Center | Design | N | Males (%) | Mean/Median (range) | Mean/Median (range) | | (mm) |
Hilton et al., 2010 | Dec 89 to Dec 09 | Single | Retrospective | 87 | 52.7 | 45 (23–76) | 48 | MCF | 12 |
Woodson et al., 2010 | 1993 to 2004 | Single | Retrospective | 49 | 61 | 48 (19–69) | 70.5 | MCF | N/A |
Vincent et al, 2012 | 1995 to 2006 | Single | Retrospective | 77 | N/A | 49 (33–68) | 102 | MCF | 9 |
Yamakami et al, 2014 | 1998 to 2012 | Single | Retrospective | 44 | 43.18 | 52+/-12 (22–69) | 81 ± 43 | RS | </= 15mm |
Quist et al, 2015 | Feb 98 to Dec 09 | Single | Retrospective | 57 | N/A | 47.65 +/- 15.11 (27–90) | 60 | MCF | N/A |
Golfinos et al, 2016 (MS Group) | 2001 to 2014 | Single | Retrospective (matched- cohort) | 85 | 45.9 | 58.0 (27.9–71.6) | HPO − 43.7; FND − 35.7 | MCF and RSA | 12.2 (median) |
Zhang et al., 2016 | Jan 1990 to Dec 2006 | Single | Retrospective | 1006 | 46.91 | 55.6 +/- 13.6 | 70.8+/-28.8 | TL, TO, RS, MCF | 17.2 (+/- 11.7) |
Ahmed et al.,2017 | 1999 to 2016 | Single | Retrospective | 155 | N/A | 49 | 84 | MCF | 9.3 |
Zhu et al, 2017 | Jan 2001 to Dec 2015 | Single | Retrospective | 110 | 51.81 | 43.2+/- 13.1 (23–70) | 49.1+/- 28.2 | RS | Max diameter </= 15mm |
Dowling et al., 2019 | 2003 to 2016 | Single | Retrospective | 43 | 40 | 45 (median) (21–64) | 37 | RS or MCF | 9 |
Table 2
Outcomes data for MS Group Studies
Author, Publication Year | Enrollment Period | Type of study | Patient's analyzed | Age (years) | F/U (months) | Type of SRS | Tumor size | Prescribed Dose |
| | Center | Design | N | Males (%) | Mean/Median (range) | Mean/Median (range) | | Volume (cc)/Max Diameter(cm) | |
Park, 2011 (GKRS group) | Jan 2004 to June 2009 | Single | Retrospective | 31 | 45.2 | 59.7 +/- 10.8 | 43.8 +/- 22.3 | GKRS | 1.93 +/- 7.05 | Maximal dose – 24.4, Marginal dose – 14.2 |
Han, 2012 | 1998 to 2009 | Single | Retrospective | 119 | 37.8 | 48 +/- 11 | 55.2 +/- 35.7 | GKRS | 1.95+/- 2.24cm3 | Mean marginal dose − 12 Gy |
Roos, 2012 | Nov 1993 to Dec 2000 | Single | Prospective | 44 | 59.1 | 63 (33 to 75) | 128 | LINAC | 2.1cm | 14 Gy (n = 4); 12 Gy (n = 40) |
Varughese, 2012 | 2000 to 2007 | Single | Prospective | 45 | 57.8 | 53.2 (R − 29 to 77.3) | 50 (23.8 to 86.7) | GKRS | 1.05cc | Mean maximum dose − 32.6 Gy |
Yomo, 2012 | July 1997 to Sept 2005 | Single | Retrospective | 154 | 50 | 54.1 (R − 24 to 78) | 60 (7 to 123) | GKRS | 0.73cc (R − 0.03 to 5.37)) | 12.1 Gy |
Carlson, 2013 | 1997 to 2002 | Single | Retrospective | 44 | 52 | 56.9 +/- 8.9 | 99.6 | GKRS | 1.698+/-2.814cc | 12-13Gy |
Sager, 2013 | July 1998 to Jan 2013 | Single | Retrospective | 51 | 41.2 | 45 (R − 20 to 77) | 51 (median) | LINAC | 1.83cm | 12Gy(median) |
Vivas, 2013 | Aug 2005 to Nov 2011 | Single | Retrospective | 63 | 44 | 59 (R − 23 to 86) | 40 (R − 6 to 81) | Cyberknife | 0.81cc | 18Gy(80% of isodose line) |
Boari, 2014 | 2001 to 2010 | Single | Retrospective | 379 | 43 | 59 (R − 23 to 85) | 75.7 | GKRS | 1.94 ± 2.2 cm3 Median 1.2 cm3 (0.013–14.3) | Maximum dose − 20-32.6(26), Marginal dose − 11–15(13) |
Carratalá, 2014 | Jan 1999 to Jan 2010 | Single | Retrospective | 35 | 49 | 58.3 (R − 31 to 76) | 56.39 | GKRS = 29; LINAC = 6 | 1.517 | 12–13 |
Combs SE, 2015 | 1990 to 2012 | Multi | Retrospective | 169 | 55 | 60 (R − 17 to 88) | 67 (R − 2 to 252) | LINAC | 1.5cm | For SRS, median dose was 13 Gy (range 10–20 Gy). |
Ikonomidis, 2015 | Jan 2002 to Aug 2010 | Single | Retrospective | 151 | 49 | 55 (R − 22 to 81) | 39 (R − 6 to 84) | LINAC | NA | 12Gy |
Golfinos, 2016 (SRS Group) | 2001 to 2014 | Single | Retrospective (matched -cohort) | 83 | 48.2 | 59.0 (R − 32 to 75) | | SRS | HPO − 1 cm; FND − 1.2 cm; TC − 1.2 cm | 12–13 |
Horiba, 2016 | Jan 2005 to Sept 2010 | Single | Retrospective | 102 | N/A | N/A | 56 (R − 24 to 99) | GKRS | NA | Mean marginal dose-11.9 Gy(11-12Gy) |
Klijn, 2016 | July 2002 to Nov 2010 | Single | Retrospective | 420 | 52 | 57.6 +/- 12.7 | 61.2 ( R − 2 to 144) | GKRS | 1.4 cc (0.01 to 17.7) | median marginal dose − 11Gy |
Watanbe, 2016 | 1990 to 2005 | Single | Retrospective | 183 | 45 | 56 (R − 11 to 80) | Imaging − 114; audiological − 59 | GKRS | 2.0cc (R − 0.05 to 26.2) | 12Gy at the periphery, 10 Gy at the facial and cochlear part |
Bowden, 2017 | Jan 2003 to Dec 2013 | Single | Retrospective | 219 | 50.2 | 55.7 (R − 23 to 99) | 49.1 | GKRS | 1.9 cm(diameter), 2.6 cm3(volume) | 12.5 (11–13) |
Milner, 2017 | 2000 to 2012 | Single | Retrospective | 45 | 45 | 55.8 (R − 34 to 75) | 69.6 | GKRS | 1.75cc | 12–13 |
Pan, 2017 | July 2003 to Dec 2008 | Single | Retrospective | 93 | 64.5 | 57.7 (+/- 15.1) | 76.3 ± 3.6 | GKRS | 3.14 ± 0.05 cc | 12 Gy |
ReuB, 2017 | 1993 to 2015 | Single | Retrospective | 49 | 57.2 | 54 (R − 20 to 77) | 65 | LINAC = 38; Cyberknife = 11 | 0.24 ± 0.12 cc | 12.6+/-0.6 Gy |
Chung, 2018 | May 2009 to Nov 2014 | Single | Retrospective | 38 | 35.7 | 65.3 (40.2 to 79.4) | 38.3 (R − 13.3 to 67.5) | LINAC | 2.13cc | 12 Gy |
Hasegawa, 2018 | May 1991 to Dec 2013 | Single | Retrospective | 92 | 41 | 54 (R − 17 to 77) | Imaging − 106; audiological − 83 | GKRS | 1.5cc | 12 Gy |
Dzierzecki, 2020 | 2011 to 2015 | Single | Retrospective | 136 | 40.4 | 54 +/- 12.6 | Imaging − 54; audiological − 41 | GKRS | 0.155cc (volume) | 12 Gy, mean cochlear dosing − 2.3 Gy |
Study characteristics
Eligible studies were published between 2010 and 2020 (as per inclusion criterion) and reported on patients treated within the period 1989 to 2016. All studies were observational and retrospective in design, except for 2 studies in SRS group which had prospective databases.[36, 42] Cases per study ranged from 43 to 1006 for the 10 studies in the MS group, with a mean of 171 patients/study and median of 71 patients. Cases per study ranged from 31 to 420 for the 23 studies in the SRS group, with a mean of 119 patients/study and median of 92 patients. In the MS group, patients were treated exclusively by the MCF approach in 5 studies;[19, 46, 43, 35, 1] exclusively by the RS approach in 2 studies;[50, 47] and by either of the two approaches (MCF or RS) in 3 studies.[16, 49, 10] In the SRS group, patients were treated exclusively with GKRS in 14 studies;[32, 17, 42, 48, 5, 3, 20, 25, 45, 4, 28, 31, 18, 11] exclusively with LINAC in 6 studies;[36, 38, 8, 22, 7, 26] exclusively with Cyberknife in 1 study;[44] with LINAC or Cyberknife in 1 study;[37] and one study did not specify the source of the SRS.[16]
Patient and Tumor characteristics
The total number of patients analyzed were 4,458; 1,713 for the MS group and 2,745 for the SRS group. The average age of patients at primary treatment ranged from 43.2 to 58 years (mean: 49.25 +/- 4.73 years) for the MS group; and from 45 to 63 (mean: 55.92 +/- 3.95 years) for the SRS group. The mean age of the MS group was significantly lower than that of the SRS group (p = 0.0002). In terms of gender composition, the percentage of males in the study cohorts ranged from 40 to 61 % (m: 48.79%) for the MS group and from 35.7–64.5% (m: 48.29%) for the SRS group (p = 0.8758). The overall mean follow-up period was 64.67 (± 20.64) months for the MS group and 66.15 (± 24.91) months for the SRS group (p = 0.8715). Table 1 and Table 2 summarize the study characteristics, patient demographic and clinical variables for patient cohorts across the 10 studies analyzed for the MS outcomes and the 23 studies analyzed for the SRS outcomes, respectively. Study outcomes are detailed in Supplementary Tables 2 and 3.
Hearing Preservation Outcome – MS versus SRS (Fig. 2)
For the MS group, 10 studies with 809 eligible patients were analyzed for HPO. Serviceable HPO was noted in 56% (CI: 52 to 59%) of cases/patient-events with a mean follow-up of 64.6 months. For the SRS group, 23 studies with 1234 eligible patients were analyzed for HPO. Serviceable HPO was noted in 59% (CI: 56 to 62%) of cases/patient-events with a mean follow-up of 66.2 months. Overall, serviceable HPO rates were noted as 56% and 59% for MS group and SRS group, respectively [p (z-test) = 0.1527] and the difference was statistically non-significant.
Tumor Control Outcome – MS versus SRS (Fig. 3)
For the MS group, 9 studies with 1635 eligible patients were analyzed for TC. TC was seen in 98% (CI: 97 to 98%) of cases with a mean follow-up of 69.8 months. For the SRS group, 19 studies with 2260 eligible patients were analyzed for TC. TC was seen in 92% (CI: 91 to 93%) of cases with a mean follow-up of 74.2 months. Overall, TC was reported in 98% and 92% of MS group patients and SRS group patients, respectively [p (z-test) < 0.0001] and the difference was statistically significant.
Facial Nerve Dysfunction – MS versus SRS (Fig. 4)
For the MS group, 8 studies with 1101 eligible patients were analyzed for FND (new onset or worsening facial nerve dysfunction >/= H & B grade 3) and FND was reported in 10% (CI: 8 to 12%) of cases. For the SRS group, 17 studies with 2285 eligible patients were analyzed for FND, which was reported in 2% (CI: 1–2%) of cases. Overall, FND rates were 10% and 2% for MS group and SRS group, respectively [p (z-test) < 0.0001] and the difference was statistically significant. For the FND outcome, there was significant heterogeneity noted for the studies in the MS arm. Amongst the outliers, the study by Zhang et al., reported FND rates of 13.6% (80 events in 585 patients);[49] while the studies by Woodson et al., Yamakami et al., and Dowling et al., reported 0, 0, and 1 event in 49, 44 and 43 patients respectively.[46, 47, 10] The studies in the SRS arm did not exhibit significant heterogeneity with respect to the FND outcome.