Table 1.
Patient characteristics
| Unadjusted total cohort | Cohort after sIPTW | |||||
|---|---|---|---|---|---|---|
| Characteristic | Palbociclib + letrozole (n = 772) | Letrozole (n = 658) | Standardized difference | Palbociclib + letrozole (n = 839) | Letrozole (n = 698) | Standardized difference |
| Age, y | ||||||
| Mean (SD) | 65.2 (10.4) | 69.2 (10.9) | − 0.3793 | 66.8 (11.2) | 67.1 (11.1) | − 0.0288 |
| Median (IQR) | 66.0 (58.0–73.0) | 70.0 (61.0–79.0) | 67.0 (59.0–75.0) | 68.0 (60.0–76.0) | ||
| Age group*, n (%), y | ||||||
| 18–49 | 52 (6.7) | 30 (4.6) | 0.0944 | 48 (5.8) | 38 (5.5) | 0.0114 |
| 50–64 | 304 (39.4) | 187 (28.4) | 0.2331 | 288 (34.3) | 244 (35.0) | − 0.0131 |
| 65–74 | 262 (33.9) | 193 (29.3) | 0.0992 | 265 (31.6) | 221 (31.7) | − 0.0025 |
| ≥ 75 | 154 (19.9) | 248 (37.7) | − 0.3995 | 237 (28.3) | 194 (27.8) | 0.0107 |
| Race/ethnicity*, n (%) | ||||||
| White | 525 (68.0) | 446 (67.8) | 0.0048 | 572 (68.1) | 470 (67.4) | 0.0163 |
| Black | 54 (7.0) | 60 (9.1) | − 0.0781 | 62 (7.4) | 55 (7.9) | − 0.0187 |
| Asian | 13 (1.7) | 10 (1.5) | 0.0131 | 15 (1.7) | 11 (1.5) | 0.0180 |
| Hispanic or Latino | 22 (2.8) | 15 (2.3) | 0.0361 | 23 (2.7) | 16 (2.3) | 0.0264 |
| Not documented† | 158 (20.5) | 127 (19.3) | 0.0292 | 167 (20.0) | 146 (20.9) | − 0.0225 |
| Practice type*, n (%) | ||||||
| Academic | 46 (6.0) | 31 (4.7) | 0.0555 | 44 (5.2) | 36 (5.2) | 0.0010 |
| Community | 726 (94.0) | 627 (95.3) | 795 (94.8) | 661 (94.8) | ||
| Disease stage at initial diagnosis*, n (%) | ||||||
| I | 82 (10.6) | 79 (12.0) | − 0.0437 | 90 (10.8) | 78 (11.2) | − 0.0127 |
| II | 187 (24.2) | 149 (22.6) | 0.0373 | 198 (23.6) | 161 (23.1) | 0.0129 |
| III | 109 (14.1) | 97 (14.7) | − 0.0177 | 121 (14.4) | 101 (14.4) | − 0.0017 |
| IV | 321 (41.6) | 254 (38.6) | 0.0608 | 338 (40.3) | 283 (40.6) | − 0.0062 |
| Not documented | 73 (9.5) | 79 (12.0) | − 0.0825 | 91 (10.9) | 74 (10.7) | 0.0069 |
| ECOG PS*, n (%) | ||||||
| 0 | 293 (38.0) | 167 (25.4) | 0.2728 | 271 (32.3) | 227 (32.5) | − 0.0059 |
| 1 | 159 (20.6) | 134 (20.4) | 0.0057 | 175 (20.9) | 144 (20.7) | 0.0055 |
| 2, 3, or 4 | 49 (6.3) | 84 (12.8) | − 0.2196 | 82 (9.7) | 66 (9.5) | 0.0095 |
| Not documented | 271 (35.1) | 273 (41.5) | − 0.1316 | 311 (37.1) | 260 (37.3) | − 0.0047 |
| Visceral disease*,‡, n (%) | ||||||
| No | 442 (57.3) | 437 (66.4) | 518 (61.8) | 429 (61.5) | ||
| Yes | 330 (42.7) | 221 (33.6) | 0.1894 | 320 (38.2) | 269 (38.5) | − 0.0060 |
| Bone-only disease*,§, n (%) | ||||||
| No | 501 (64.9) | 398 (60.5) | 530 (63.1) | 440 (63.1) | ||
| Yes | 271 (35.1) | 260 (39.5) | − 0.0913 | 309 (36.9) | 258 (36.9) | − 0.0011 |
| Brain metastases, n (%) | ||||||
| No | 753 (97.5) | 628 (95.4) | 820 (97.7) | 656 (94.0) | ||
| Yes | 19 (2.5) | 30 (4.6) | − 0.1142 | 19 (2.3) | 42 (6.0) | − 0.1901 |
| Time from initial Dx to metastatic Dx*, n (%), y | ||||||
| De novo | 321 (41.6) | 254 (38.6) | 0.0608 | 338 (40.3) | 283 (40.6) | − 0.0062 |
| ≤ 1 | 19 (2.5) | 23 (3.5) | − 0.0609 | 25 (2.9) | 21 (3.1) | − 0.0070 |
| > 1–5 | 123 (15.9) | 111 (16.9) | − 0.0253 | 133 (15.9) | 111 (16.0) | − 0.0016 |
| > 5 | 308 (39.9) | 269 (40.9) | − 0.0201 | 342 (40.7) | 281 (40.3) | 0.0097 |
| Not documented | 1 (0.1) | 1 (0.2) | − 0.0060 | 1 (0.1) | 1 (0.1) | 0.0010 |
| Number of metastatic sites*,ǁ, n (%) | ||||||
| 1 | 372 (48.2) | 365 (55.5) | − 0.1462 | 423 (50.4) | 355 (50.9) | − 0.0104 |
| 2 | 220 (28.5) | 147 (22.3) | 0.1418 | 217 (25.9) | 183 (26.2) | − 0.0070 |
| 3 | 110 (14.2) | 64 (9.7) | 0.1396 | 101 (12.0) | 86 (12.4) | − 0.0104 |
| 4 | 40 (5.2) | 18 (2.7) | 0.1257 | 35 (4.2) | 27 (3.8) | 0.0184 |
| ≥ 5 | 20 (2.6) | 10 (1.5) | 0.0755 | 18 (2.2) | 15 (2.2) | 0.0002 |
| Not documented | 10 (1.3) | 54 (8.2) | −0.3293 | 44 (5.3) | 31 (4.5) | 0.0371 |
The balance in important prognostic baseline characteristics was assessed using a standardized difference approach, with a standardized difference of ≥ 0.10 considered indicative of practical significance [24]. The total patient population for different subgroups varied due to the application of sIPTW. Therefore, the total n number for each subgroup may not have always equaled the N number of the treatment arm (due to rounding error and categorization differences). Calculated percentages were based on the number of patients reported within each subgroup
Dx diagnosis, ECOG PS Eastern Cooperative Oncology Group performance status, IQR interquartile range, sIPTW stabilized inverse probability treatment weighting
*Variable used in the propensity score matching model; de novo vs not de novo were used as categories for initial Dx to metastatic Dx
†Race data were not known in the “not documented” race group
‡Visceral disease was defined as metastatic disease in the lung and/or liver; patients could have had other sites of metastases. No visceral disease was defined as no lung or liver metastases
§Bone-only disease was defined as metastatic disease in the bone only
ǁMultiple metastases at the same site were counted as 1 site (e.g., if a patient had 3 bone metastases in the spine, it was considered only 1 site)