Cox regression results for survival comparisons
| Treatment (data set) . | Comparison . | n1 . | n2 . | Median survival 1, mo . | Median survival 2, mo . | HR* (CI) . | P . |
|---|---|---|---|---|---|---|---|
| Bortezomib based (C) | Bortezomib standard vs bortezomib good (PFS) | 108 | 39 | 21.9 | 36.2 | 0.37 (0.17-0.81) | .0063 |
| RD (C) | Bortezomib standard vs bortezomib good (PFS) | 29 | 11 | Not reached | 18.8 | 2.35 (0.76-7.63) | .16 |
| Bortezomib based or RD (C) | Incorrectly treated vs correctly treated (PFS) | 119 | 68 | 20.1 | Not reached | 0.40 (0.23-0.72) | .0012 |
| Bortezomib based or RD (C) | Incorrectly treated vs correctly treated (OS) | 119 | 68 | 30.7 | Not reached | 0.41 (0.21-0.80) | .0049 |
| Bortezomib based or RD or VRD (C) | All VRD vs all non-VRD (PFS) | 208 | 187 | 43.7 | 26.0 | 1.50 (1.06-2.13) | .02 |
| Bortezomib based or RD or VRD (C) | All VRD vs all non-VRD (OS) | 208 | 187 | Not reached | 37.8 | 2.05 (1.34-3.13) | .00084 |
| Bortezomib based or RD or VRD (C) | All VRD vs non-VRD correctly treated (OS) | 208 | 68 | Not reached | Not reached | 1.12 (0.58-2.13) | .74 |
| Bortezomib based or RD or VRD (C) | All VRD vs non-VRD correctly treated (PFS) | 208 | 68 | 43.7 | Not reached | 0.86 (0.50-1.47) | .57 |
| Bortezomib based ± ASCT | Bortezomib correctly treated vs ASCT (PFS) | 47 | 39 | 36.6 | 36.2 | 1.16 (0.47-2.87) | .75 |
| Bortezomib based ± ASCT | Bortezomib correctly treated vs ASCT (OS) | 47 | 39 | 43.7 | Not reached | 3.10 (0.80-12.01) | .08 |
| VRD (C) | Lenalidomide best vs bortezomib best (OS) | 164 | 44 | Not reached | Not reached | 1.41 (0.69-2.90) | .36 |
| VRD (C) | Lenalidomide best vs bortezomib best (PFS) | 164 | 44 | 43.7 | 45.4 | 1.11 (0.62-1.97) | .74 |
| Bortezomib alone (M) | Lenalidomide best vs bortezomib best (PFS) | 135 | 38 | 4.14 | 4.77 | 0.66 (0.43-1.0) | .04 |
| Bortezomib alone (M) | Lenalidomide best vs bortezomib best (OS) | 148 | 40 | 15.2 | 21.8 | 0.57 (0.53-0.91) | .01 |
| RD (PCL) | Lenalidomide best vs bortezomib best (PFS) | 8 | 10 | Not reached | 1.0 | Not defined† | 2.52 × 10−5 |
| RD (PCL) | Lenalidomide best vs bortezomib best (OS) | 8 | 10 | Not reached | 12.5 | Not defined† | .0013 |
| PAD/ASCT (H) | Lenalidomide best vs bortezomib best (PFS) | 84 | 57 | 26.6 | 31.8 | 0.91 (0.60-1.37) | .63 |
| PAD/ASCT (H) | Lenalidomide best vs bortezomib best (OS) | 85 | 58 | Not reached | Not reached | 0.73 (0.40-1.34) | .30 |
| Treatment (data set) . | Comparison . | n1 . | n2 . | Median survival 1, mo . | Median survival 2, mo . | HR* (CI) . | P . |
|---|---|---|---|---|---|---|---|
| Bortezomib based (C) | Bortezomib standard vs bortezomib good (PFS) | 108 | 39 | 21.9 | 36.2 | 0.37 (0.17-0.81) | .0063 |
| RD (C) | Bortezomib standard vs bortezomib good (PFS) | 29 | 11 | Not reached | 18.8 | 2.35 (0.76-7.63) | .16 |
| Bortezomib based or RD (C) | Incorrectly treated vs correctly treated (PFS) | 119 | 68 | 20.1 | Not reached | 0.40 (0.23-0.72) | .0012 |
| Bortezomib based or RD (C) | Incorrectly treated vs correctly treated (OS) | 119 | 68 | 30.7 | Not reached | 0.41 (0.21-0.80) | .0049 |
| Bortezomib based or RD or VRD (C) | All VRD vs all non-VRD (PFS) | 208 | 187 | 43.7 | 26.0 | 1.50 (1.06-2.13) | .02 |
| Bortezomib based or RD or VRD (C) | All VRD vs all non-VRD (OS) | 208 | 187 | Not reached | 37.8 | 2.05 (1.34-3.13) | .00084 |
| Bortezomib based or RD or VRD (C) | All VRD vs non-VRD correctly treated (OS) | 208 | 68 | Not reached | Not reached | 1.12 (0.58-2.13) | .74 |
| Bortezomib based or RD or VRD (C) | All VRD vs non-VRD correctly treated (PFS) | 208 | 68 | 43.7 | Not reached | 0.86 (0.50-1.47) | .57 |
| Bortezomib based ± ASCT | Bortezomib correctly treated vs ASCT (PFS) | 47 | 39 | 36.6 | 36.2 | 1.16 (0.47-2.87) | .75 |
| Bortezomib based ± ASCT | Bortezomib correctly treated vs ASCT (OS) | 47 | 39 | 43.7 | Not reached | 3.10 (0.80-12.01) | .08 |
| VRD (C) | Lenalidomide best vs bortezomib best (OS) | 164 | 44 | Not reached | Not reached | 1.41 (0.69-2.90) | .36 |
| VRD (C) | Lenalidomide best vs bortezomib best (PFS) | 164 | 44 | 43.7 | 45.4 | 1.11 (0.62-1.97) | .74 |
| Bortezomib alone (M) | Lenalidomide best vs bortezomib best (PFS) | 135 | 38 | 4.14 | 4.77 | 0.66 (0.43-1.0) | .04 |
| Bortezomib alone (M) | Lenalidomide best vs bortezomib best (OS) | 148 | 40 | 15.2 | 21.8 | 0.57 (0.53-0.91) | .01 |
| RD (PCL) | Lenalidomide best vs bortezomib best (PFS) | 8 | 10 | Not reached | 1.0 | Not defined† | 2.52 × 10−5 |
| RD (PCL) | Lenalidomide best vs bortezomib best (OS) | 8 | 10 | Not reached | 12.5 | Not defined† | .0013 |
| PAD/ASCT (H) | Lenalidomide best vs bortezomib best (PFS) | 84 | 57 | 26.6 | 31.8 | 0.91 (0.60-1.37) | .63 |
| PAD/ASCT (H) | Lenalidomide best vs bortezomib best (OS) | 85 | 58 | Not reached | Not reached | 0.73 (0.40-1.34) | .30 |
All signature assignments were based on the 7-gene signature. Correctly treated patients (rows 3, 4, and 7-10) were those predicted by the signature as lenalidomide best and who were treated with RD or those predicted as bortezomib best and treated with bortezomib-based therapy.
C, CoMMpass data set; CI, confidence interval; H, HOVON/GMMG data set; M, Millennium data set; n1, first group in comparison; n2, second group in comparison; OS, overall survival.
HR: second group vs first group.
HR for the PCL group is not defined because all patients in group 2 experienced progression before any progressions in group 1 (row 11) or because there were no deaths in group 1 (row 12).