Table 1 Clinical feature distributions for number of BMs, BMs progressing post-SRS, and patients (where applicable) for the study sample.

From: Performance sensitivity analysis of brain metastasis stereotactic radiosurgery outcome prediction using MRI radiomics

Clinical features

# Patients

# BMs (% progression)

Sex (p = 0.831)

Female

55

66 (21.2%)

Male

44

57 (22.8%)

Age (p = 0.925)

Median (range)

58.0 (38.4–86.0) years

Primary cancer active (p = 0.002)

Yes

44

55 (9.1%)

No

55

68 (32.4%)

Primary cancer site (p = 0.003)

Lung

59

70 (12.9%)

Breast

10

14 (35.7%)

Renal

10

15 (13.3%)

Colorectal

8

10 (40.0%)

Skin

8

9 (66.7%)

Other

4

5 (20.0%)

Primary cancer histology (p = 0.001)

Adenocarcinoma

49

65 (20.0%)

NSCLC

31

36 (11.1%)

Melanoma

8

9 (66.7%)

Squamous carcinoma

7

8 (50.0%)

Other

4

5 (0.0%)

Extracranial systemic metastases (p = 0.665)

Yes

39

50 (20.0%)

No

60

73 (23.3%)

Systemic therapy status (p = 0.033)

Radical

51

10 (20.0%)

Palliative

41

60 (31.7%)

None

7

53 (11.3%)

Neurological symptoms steroid response (p = 0.426)

Fully resolved

48

31 (16.1%)

Improvement

7

4 (50.0%)

Limited improvement

4

56 (25.0%)

No improvement

26

11 (9.1%)

Unknown

14

21 (23.8%)

ECOG/WHO performance score (p = 0.580)

0

31

39 (17.9%)

1

60

73 (21.9%)

2

6

9 (33.3%)

3

2

2 (50.0%)

GTV volume (p < 0.001)

Median (range)

3.07 (0.02–30.23) cc

< 7.5 cc

94 (17.0%)

> 7.5 cc

29 (37.9%)

BM location (p = 0.626)

Supratentorial

96 (22.9%)

Infratentorial

27 (18.5%)

SRS prescription (p = 0.003)

15 Gy in 1 fraction

5 (0.0%)

18 Gy in 1 fraction

36 (30.6%)

21 Gy in 1 fraction

72 (13.9%)

24 Gy in 3 fractions

10 (60.0%)

  1. The “Neurological Symptoms Steroid Response” feature qualitatively scores the improvement of neurological symptoms after the administration of steroids, based on the previously reported methodology of Lagerwaard et al.24. p-values provided for statistical comparisons between BMs that progressed and did not progress post-SRS. The Wilcoxon rank sum test was used for continuous features (age and GTV volume). The Chi-squared test was used for the remaining categorical features. NSCLC non-small cell lung cancer, ECOG Eastern Cooperative Oncology Group, WHO World Health Organization.