Abstract
Purpose:
Brain metastases are a common problem in patients with melanoma, but little is known about the effect of gene mutations on survival in these patients.
Methods and Materials:
We created a retrospective multi-institutional database of 823 patients with melanoma and brain metastases diagnosed between 2006 and 2015. Clinical parameters, gene mutation status (BRAF, C-KIT, NRAS ), and treatment were correlated with survival. Treatment patterns and outcomes were compared with a prior era (1985–2005).
Results:
BRAF status was known in 584 of 823 patients (71%). BRAF, NRAS, and C-KIT mutations were present in 51%, 22%, and 11% of tested patients, respectively. The median time from primary diagnosis to brain metastasis was 32 months, and overall median survival (MS) from the time of initial treatment of brain metastases was 10 months. MS for BRAF-positive and BRAF-negative patients was 13 months and 9 months, respectively (P = .02). There was no significant difference in MS in patients with or without NRAS or C-KIT mutations. The time from primary diagnosis to brain metastasis did not vary by mutation and was not associated with survival after the diagnosis of brain metastases. MS for the 1985 to 2005 and 2006 to 2015 cohorts was 6.7 months and 10.0 months, respectively (P<.01). Reflecting treatment-trend changes, use of whole-brain radiation therapy decreased from 48% to 26% during this period. Among BRAF-positive patients, 71% received targeted BRAF and/or MEK inhibitors and 57% received some combination of targeted therapy, chemotherapy, and/ or immunotherapy.
Conclusions:
For melanoma patients with brain metastases, BRAF-positive patients survive longer than BRAF-negative patients and overall survival has improved from 1985–2005 to 2006–2015.
Summary
This retrospective study of gene mutations in 823 melanoma patients with brain metastases shows that BRAF- positive patients survive longer than BRAF-negative patients after the diagnosis of brain metastases and that overall survival for these patients receiving diagnoses from 2006 to 2015 is improved compared with 1985 to 2005.
Introduction
The management of metastatic melanoma is rapidly evolving. Recent landmark trials have shown a survival benefit for immunotherapy (both CTLA-4 and PD-1 inhibition, independently and in combination) in selected patients (1–3), as well as for targeted therapies (both BRAF and MEK inhibitors, independently and in combination) (4–6). Despite these advances, brain metastases remain a common cause of morbidity and death in melanoma patients. In 2016, an estimated 76,380 patients were diagnosed with melanoma and approximately 10,000 will die from the disease (7). In nearly half of all melanoma patients, brain metastases will develop at some point in the course of their disease (8, 9), and brain metastases are the cause of death in 20% to 54% of patients with melanoma (10). Although melanoma represents only 4% of all cancers, it has garnered intense interest because of the progress achieved with targeted therapies and immunotherapies. Multiple preclinical studies (11–13) and case reports (14–16) have described radiation-induced immune enhancement (abscopal effect). Melanoma is also of interest because it has the highest propensity of all cancers to metastasize to the brain, and the underlying biological susceptibility for this is ill understood (17).
We previously demonstrated that the survival of patients with brain metastases varies widely by diagnosis and diagnosis-specific prognostic factors as defined by the Diagnosis-Specific Graded Prognostic Assessment (GPA) (18, 19). The melanoma cohort in our original study (1985–2005, n = 481) had a median survival of 6.7 months from the time of initial brain metastasis treatment. The only significant prognostic factors for survival in the study were Karnofsky Performance Status (KPS) and number of brain metastases. Our group recently reported prolonged survival in lung adenocarcinoma patients with EGFR and ALK alterations (20), and the 2010 lung GPA was updated accordingly (21). Little is known about the impact of gene mutations and the aforementioned systemic therapies on prognosis for melanoma patients with brain metastases. The purpose of this study was to determine the effect of gene mutations on survival and the time from primary diagnosis to brain metastasis (TPDBM).
Methods and Materials
We created a multi-institutional institutional review board—approved retrospective database of 823 patients with melanoma and brain metastases diagnosed between 2006 and 2015, nonoverlapping with our earlier cohort. Clinical parameters, gene mutation status (BRAF, C-KIT, NRAS ), and treatment were recorded, and each was analyzed for association with survival (measured from time of initiation of treatment of brain metastases), TPDBM, and cause of death. The log-rank test and Wilcoxon rank sum test were used to compare median survival and TPDBM, respectively. Multivariate Cox regression was used to confirm that noted survival differences are independent of other prognostic factors in the GPA. Data regarding the source of the tissue (primary vs brain) used for mutation assessment were not collected. The mutation status was determined by a variety of different Clinical Laboratory Improvement Amendments (CLIA)-approved laboratory methods used in the 10 participating academic institutions.
Results
Patient characteristics
Table 1 shows the patient characteristics. Salient observations include that two-thirds of patients were male patients, two-thirds of patients died of nonneurologic causes (as reported in the medical record or deduced on retrospective chart review), younger patients were more likely to be BRAF positive, and survival varied directly with KPS and indirectly with the number of brain metastases and the presence of extracranial metastases. Survival from the time of brain metastasis diagnosis did not correlate with the TPDBM. The 2010 melanoma GPA was again confirmed as an accurate tool to estimate survival.
Table 1.
Patient characteristics by mutation status
| BRAF | NRAS | C-KIT | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Overall | Positive | Negative | Unknown | Positive | Negative | Unknown | Positive | Negative | Unknown | ||||||||||||
| Variable | Category | n | MS, mo | n (%) | MS, mo | n (%) | MS, mo | n (%) | MS, mo | n(%) | MS, mo | n (%) | MS, mo | n (%) | MS, mo | n(%) | MS, mo | n(%) | MS, mo | n (%) | MS, mo |
| Overall | 823 | 10 | 297 (36) | 13 | 287 (35) | 9 | 239 (29) | 8 | 63 (8) | 14 | 222 (27) | 11 | 538 (65) | 9 | 32(4) | 9 | 261 (32) | 11 | 530 (64) | 10 | |
| Melanoma GPA | 0–1 | 159 | 6 | 66 (42) | 8 | 53 (33) | 4 | 40 (25) | 5 | 8(5) | 3 | 39 (25) | 7 | 112 (70) | 6 | 7(4) | 8 | 48 (30) | 6 | 104 (65) | 6 |
| 2 | 220 | 9 | 83 (38) | 11 | 76 (35) | 9 | 61 (28) | 6 | 12(5) | 6 | 65 (30) | 12 | 143 (65) | 7 | 7(3) | 8 | 76 (35) | 9 | 137 (62) | 8 | |
| 3 | 239 | 10 | 78 (33) | 15 | 82 (34) | 9 | 79 (33) | 8 | 17 (7) | 32 | 67 (28) | 12 | 155 (65) | 9 | 11 (5) | 10 | 71 (30) | 13 | 157 (66) | 10 | |
| 4 | 205 | 17 | 70 (34) | 22 | 76 (37) | 18 | 59 (29) | 12 | 26 (13) | 22 | 51 (25) | 18 | 128 (62) | 15 | 7(3) | 10 | 66 (32) | 19 | 132 (64) | 15 | |
| Cause of death | Neurologic | 181 | 8 | 67 (37) | 10 | 66 (36) | 7 | 48 (27) | 8 | 16(9) | 8 | 31 (17) | 10 | 134 (74) | 7 | 6(3) | 14 | 41 (23) | 10 | 134 (74) | 7 |
| Nonneurologic/cancer | 296 | 7 | 100 (34) | 10 | 116 (39) | 7 | 80 (27) | 6 | 24(8) | 9 | 114 (39) | 8 | 158 (53) | 7 | 14(5) | 7 | 119 (40) | 7 | 163 (55) | 7 | |
| Nonneurologic/noncancer | 8 | 8 | 2(25) | 10 | 2(25) | 10 | 4(50) | 6 | 24(8) | 9 | 2(25) | 9 | 6(75) | 8 | 14(5) | 7 | 2(25) | 9 | 6(75) | 8 | |
| Unknown | 164 | 8 | 49 (30) | 9 | 40 (24) | 5 | 75 (46) | 8 | 7(4) | 7 | 16 (10) | 9 | 141 (86) | 8 | 5(3) | 7 | 34 (21) | 9 | 125 (76) | 8 | |
| Time from PD to BM | 0–12 mo | 188 | 9 | 70 (37) | 12 | 68 (36) | 9 | 50 (27) | 5 | 14(7) | 13 | 49 (26) | 10 | 125 (66) | 8 | 7(4) | 9 | 66 (35) | 12 | 115 (61) | 8 |
| 12–30 mo | 199 | 8 | 62 (31) | 10 | 75 (38) | 8 | 62 (31) | 8 | 14(7) | 18 | 51 (26) | 8 | 134 (67) | 8 | 11 (6) | 7 | 59 (30) | 10 | 129 (65) | 8 | |
| 30–60 mo | 216 | 10 | 84 (39) | 14 | 80 (37) | 9 | 52 (24) | 10 | 19(9) | 7 | 66 (31) | 14 | 131 (61) | 10 | 7(3) | 14 | 68 (31) | 12 | 141 (65) | 10 | |
| >60 mo | 220 | 11 | 81 (37) | 16 | 64 (29) | 11 | 75 (34) | 8 | 16(7) | 17 | 56 (25) | 11 | 148 (67) | 10 | 7(3) | 28 | 68 (31) | 11 | 145 (66) | 11 | |
| Age at BM diagnosis | <50 y | 192 | 15 | 98 (51) | 15 | 39 (20) | 11 | 55 (29) | 14 | 11 (6) | 34 | 55 (29) | 14 | 126 (66) | 14 | 10(5) | - | 56 (29) | 15 | 126 (66) | 14 |
| 50–59 y | 221 | 9 | 76 (34) | 11 | 71 (32) | 12 | 74 (33) | 8 | 17 (8) | 17 | 53 (24) | 12 | 151 (68) | 8 | 8(4) | 8 | 63 (29) | 13 | 150 (68) | 9 | |
| 60–69 y | 238 | 10 | 76 (32) | 12 | 105 (44) | 9 | 57 (24) | 9 | 23 (10) | 13 | 59 (25) | 13 | 156 (66) | 9 | 8(3) | 9 | 82 (34) | 11 | 148 (62) | 9 | |
| 70–97 y | 172 | 7 | 47 (27) | 13 | 72 (42) | 7 | 53 (31) | 5 | 12(7) | 4 | 55 (32) | 10 | 105 (61) | 7 | 6(3) | 9 | 60 (35) | 8 | 106 (62) | 7 | |
| Sex | Male | 541 | 9 | 183 (34) | 11 | 207 (38) | 9 | 151 (28) | 8 | 43 (8) | 17 | 166 (31) | 10 | 332 (61) | 8 | 25 (5) | 8 | 185 (34) | 10 | 331 (61) | 9 |
| Female | 281 | 11 | 114 (41) | 15 | 79 (28) | 10 | 88 (31) | 9 | 20 (7) | 13 | 56 (20) | 14 | 205 (73) | 11 | 7(2) | 11 | 76 (27) | 13 | 198 (70) | 11 | |
| KPS | <70 | 76 | 7 | 29 (38) | 8 | 26 (34) | 5 | 21 (28) | 4 | 1 (1) | 2 | 14 (18) | 10 | 61 (80) | 5 | 3(4) | 8 | 20 (26) | 10 | 53 (70) | 5 |
| 70 | 102 | 6 | 34 (33) | 7 | 39 (38) | 7 | 29 (28) | 4 | 4(4) | 7 | 19 (19) | 5 | 79 (77) | 6 | 2(2) | 7 | 26 (25) | 7 | 74 (73) | 6 | |
| 80 | 192 | 8 | 72 (38) | 12 | 58 (30) | 6 | 62 (32) | 8 | 13(7) | 6 | 46 (24) | 11 | 133 (69) | 8 | 5(3) | 9 | 55 (29) | 8 | 132 (69) | 8 | |
| 90 | 343 | 12 | 124 (36) | 15 | 118 (34) | 14 | 101 (29) | 9 | 34 (10) | 20 | 111 (32) | 13 | 198 (58) | 11 | 13(4) | 8 | 122 (36) | 13 | 208 (61) | 12 | |
| 100 | 110 | 17 | 38 (35) | 22 | 46 (42) | 17 | 26 (24) | 10 | 11 (10) | 49 | 32 (29) | 15 | 67 (61) | 16 | 9(8) | 28 | 38 (35) | 17 | 63 (57) | 15 | |
| No. of brain metastases | 1 | 328 | 12 | 112 (34) | 15 | 111 (34) | 14 | 105 (32) | 10 | 30 (9) | 20 | 72 (22) | 19 | 226 (69) | 11 | 10(3) | 9 | 88 (27) | 20 | 230 (70) | 10 |
| 2 | 151 | 10 | 52 (34) | 16 | 54 (36) | 10 | 45 (30) | 8 | 11 (7) | 20 | 48 (32) | 13 | 92 (61) | 9 | 7(5) | 25 | 56 (37) | 10 | 88 (58) | 11 | |
| 3–5 | 190 | 8 | 56 (29) | 14 | 75 (39) | 7 | 59 (31) | 8 | 11 (6) | 29 | 44 (23) | 9 | 135 (71) | 7 | 8(4) | 11 | 60 (32) | 7 | 122 (64) | 9 | |
| >5 | 154 | 7 | 77 (50) | 9 | 47 (31) | 5 | 30 (19) | 5 | 11 (7) | 3 | 58 (38) | 8 | 85 (55) | 7 | 7(5) | 7 | 57 (37) | 8 | 90 (58) | 7 | |
| Extracranial metastases | No | 136 | 20 | 47 (35) | 35 | 41 (30) | 24 | 48 (35) | 10 | 15(11) | 27 | 34 (25) | 46 | 87 (64) | 12 | 5(4) | 8 | 50 (37) | 34 | 81 (60) | 16 |
| Yes | 687 | 9 | 250 (36) | 11 | 246 (36) | 8 | 191 (28) | 8 | 48 (7) | 12 | 188 (27) | 10 | 451 (66) | 8 | 27(4) | 10 | 211 (31) | 10 | 449 (65) | 9 | |
Abbreviations: BM = brain metastasis; GPA = Graded Prognostic Assessment; KPS = Karnofsky Performance Status; MS = median survival (Kaplan-Meier estimate); PD = primary diagnosis.
Percentages are based on the overall number in each row.
Survival and TPDBM
Table 2 shows median survival and TPDBM by gene status. Among patients tested for gene status, BRAF, NRAS, and C-KIT mutations were present in 297 of 584 patients (51%), 63 of 285 patients (22%), and 32 of 293 patients (11%), respectively. Only 2% of patients showed positive results for multiple mutations. Unlike lung cancer, there was no strong interaction among these mutations with respect to overall survival, so each was analyzed separately. Median TPDBM was 32 months, and median overall survival was 10 months from the time of first treatment of brain metastasis. Patients with known BRAF status survived longer than those with unknown BRAF status (11 months [interquartile range (IQR), 5–30 months] vs 8 months [IQR, 4–17 months]; P = .0007). Notably, most of the patients with unknown BRAF status received diagnoses in 2006 to 2009 before BRAF testing became routine, and conversely, most of the patients with known BRAF status received diagnoses from 2010 to 2015. Vemurafenib, the first BRAF- targeted therapy, was approved by the Food and Drug Administration on August 17, 2011. The difference in the survival rate between known and unknown BRAF status correlates with the steady improvement in survival over time. Similarly, patients with known NRAS status survived longer than those with unknown NRAS status (12 months [IQR, 6–30 months] vs 9 months [IQR, 4–21 months]; P = .002). There was no significant survival difference between patients with known and unknown C-KIT status. BRAF-positive patients survived longer than BRAF-negative patients (13 months [IQR, 6–33 months] vs 9 months [IQR, 5–24 months]; P = .02) (Fig. 1). Furthermore, after we adjusted for the existing melanoma GPA, BRAF-positive patients had superior overall survival compared with BRAF-negative patients (hazard ratio, 0.74; 95% confidence interval, 0.61–0.90; P<.01). There was no significant survival difference between NRAS-positive and NRAS-negative patients (14 months [IQR, 5–36 months] vs 11 months [IQR, 6–32 months]; P = .79) or between C-KIT—positive and C-KIT—negative patients (9 months [IQR, 7–25 months] vs 11 months [IQR, 5–30 months]; P = .95). Regarding TPDBM, there was no significant difference between known and unknown or positive and negative gene status for BRAF, NRAS, or C-KIT.
Table 2.
MS and TPDBM by gene status
| Survival | TPDBM | |||||||
|---|---|---|---|---|---|---|---|---|
| Gene | Status | n(%) | Median, mo | IQR, mo | P value | Median, mo | IQR, mo | P value |
| Overall | 823 | 10 | 4–26 | 32 | 14–63 | |||
| BRAF | Unknown | 239 (29) | 8 | 4–17 | 33 | 15–72 | ||
| Known | 584 (71) | 11 | 5–30 | .0007*† | 32 | 14–59 | .34†‡ | |
| Negative | 287 (35) | 9 | 5–24 | 30 | 13–54 | |||
| Positive | 297 (36) | 13 | 6–33 | .02*§ | 35 | 14–65 | .11‡§ | |
| NRAS | Unknown | 538 (65) | 9 | 4–21 | 32 | 13–65 | ||
| Known | 285 (35) | 12 | 6–30 | .002*† | 34 | 14–60 | .92†‡ | |
| Negative | 222 (27) | 11 | 6–32 | 34 | 15–62 | |||
| Positive | 63 (8) | 14 | 5–36 | 79*§ | 35 | 14–60 | .82‡§ | |
| C-KIT | Unknown | 530 (64) | 10 | 4–25 | 33 | 15–64 | ||
| Known | 293 (36) | 10 | 5–28 | .09*† | 31 | 12–62 | .15†‡ | |
| Negative | 261 (32) | 11 | 5–30 | 32 | 12–62 | |||
| Positive | 32(4) | 9 | 7–25 | .95*§ | 29 | 14–43 | .58‡§ | |
Abbreviations: IQR = interquartile range (25th percentile-75th percentile); TPDBM = time from primary diagnosis to brain metastasis. Percentages are based on the total number of patients (N = 823). All tests were performed within each gene.
P values are from log-rank tests comparing survival distributions between groups.
P values are from comparison of the group with known gene alteration status and the group with unknown status.
P values are from Wilcoxon rank sum tests comparing TPDBM between groups.
P values are from comparison of the group with positive gene alteration and the group with negative gene alteration.
Fig. 1.
Kaplan-Meier curve of survival by BRAF mutation status. Abbreviations: BM = brain metastasis; MS = median survival.
Treatment
Table 3 shows an analysis of median survival and risk of death (hazard ratio) by treatment and by treatment era, adjusted by GPA. Salient observations include the following: (1) Use of whole-brain radiation therapy (WBRT) decreased from 48% to 26% between the 2 treatment periods (from 1985–2005 to 2006–2015); and (2) use of stereotactic radiosurgery (SRS) (76% and 78%) and surgery (18% and 17%) remained nearly the same. Chemotherapy data were not available for the earlier study period, and that period predated the use of targeted therapy and immunotherapy.
Table 3.
Median survival and risk of death by treatment and by treatment era
| Overall | WBRT | SRS | WBRT + SRS |
S + SRS | S + WBRT | S + WBRT + SRS |
S | |
|---|---|---|---|---|---|---|---|---|
| Historical cohort | ||||||||
| n | 481 | 86 (18%) | 221 (46%) | 89 (18%) | 30 (6%) | 29 (6%) | 26 (5%) | |
| Median survival, mo | 7 | 3 | 7 | 7 | 13 | 11 | 13 | |
| Mean GPA | 2.5 | 1.9 | 2.7 | 2.4 | 2.9 | 3.1 | 2.7 | |
| Risk of death (HR) | 1.00 | 0.74 | 0.83 | 0.76 | 0.60 | 0.49 | ||
| 95% Cl | 0.51–1.07 | 0.57–1.21 | 0.45–1.29 | 0.38–0.99 | 0.28–0.86 | |||
| P value | .11 | .33 | .30 | .04 | .01 | |||
| Current study | ||||||||
| n | 823 | 91 (11%) | 464 (56%) | 73 (9%) | 95 (12%) | 34 (4%) | 12 (1%) | 43 (5%) |
| Median survival, mo | 10 | 6 | 10 | 9 | 13 | 11 | 11 | 22 |
| Mean GPA | 2.6 | 1.7 | 2.8 | 2.1 | 2.6 | 1.5 | 2.4 | 2.8 |
| Risk of death (HR) | 1.00 | 0.69 | 0.62 | 0.50 | 0.54 | 0.70 | 0.42 | |
| 95% Cl | 0.54–0.89 | 0.45–0.86 | 0.36–0.69 | 0.35–0.84 | 0.36–1.36 | 0.27–0.63 | ||
| P value | <.01 | <.01 | <.01 | <.01 | .29 | <.01 | ||
Abbreviations: Cl = confidence interval; GPA = Graded Prognostic Assessment; HR = hazard ratio; S = surgery; SRS = stereotactic radiosurgery; WBRT = whole-brain radiation therapy.
HR, 95% Cl, and P values (each treatment vs WBRT alone within each cohort) are adjusted for GPA. Median survival is unadjusted. Eleven patients in this study had no treatment.
Chemotherapies, targeted therapies, and immunotherapies
Among BRAF-positive patients, 194 of 272 (71%) received BRAF and/or MEK targeted drugs and 156 of 272 (57%) received some combination of BRAF and/or MEK targeted drugs with either immunotherapy or chemotherapy. Immunotherapy use increased from 40% to 50% to over 70% between 2009 and 2011, whereas chemotherapy use declined from 70% in 2010 to 28% in 2014. From 2006 to 2015, almost half of all patients (405 of 823 [49%]) received chemotherapy (carboplatin [n = 228], paclitaxel [n = 99], temozolomide [n = 262], bevacizumab [n = 29], investigational agents [n = 18], other [n=105]). Regarding the timing of chemotherapy, nearly equal numbers of patients received chemotherapy before (228 of 405 [28%]) and after (239 of 405 [29%]) the diagnosis of brain metastases. More than half (224 of 405 [55%]) had complete data in terms of the start and stop dates of chemotherapy. The median duration of chemotherapy was 2 months. Table 4 shows the number of patients receiving immunotherapy, targeted therapy, and chemotherapy by year and BRAF status.
Table 4.
Number of patients receiving immunotherapy, targeted therapy, or chemotherapy by year and BRAF status
| 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | |
|---|---|---|---|---|---|---|---|---|---|---|
| Immunotherapy/all patients (%) | 16/39 (41) | 17/38 (45) | 42/75 (56) | 29/66 (44) | 53/86 (62) | 88/121 (73) | 114/155 (74) | 94/130 (72) | 71/96 (74) | 11/17 (65) |
| Immunotherapy/BRAF | 2/2 (100) | 1/1 (100) | 5/8 (63) | 12/22 (55) | 16/30 (53) | 29/49 (59) | 47/71 (66) | 47/70 (67) | 23/36 (64) | 5/8 (63) |
| positive (%) | ||||||||||
| Immunotherapy/BRAF | 0/1 (0) | 1/2 (50) | 6/8 (75) | 5/13 (38) | 18/30 (60) | 48/55 (87) | 62/72 (86) | 43/53 (81) | 41/46 (89) | 4/7 (57) |
| negative (%) | ||||||||||
| Immunotherapy/BRAF | 14/36 (39) | 15/35 (43) | 31/59 (53) | 12/31 (39) | 19/26 (73) | 11/17 (65) | 5/12 (42) | 4/7 (57) | 7/14 (50) | 2/2 (100) |
| untested (%) | ||||||||||
| Targeted therapy/all patients (%) | 6/39 (15) | 6/38 (16) | 11/75 (15) | 13/66 (20) | 26/86 (30) | 47/121 (39) | 65/155 (42) | 58/130 (45) | 39/96 (41) | 6/17 (35) |
| Targeted therapy/BRAF | 2/2 (100) | 1/1 (100) | 5/8 (63) | 9/22 (41) | 16/30 (53) | 39/49 (80) | 57/71 (80) | 52/70 (74) | 29/36 (81) | 5/8 (63) |
| positive (%) | ||||||||||
| Targeted therapy/BRAF | 0/1 (0) | 0/2 (0) | 2/8 (25) | 3/13 (23) | 5/30 (17) | 7/55 (13) | 5/72 (7) | 3/53 (6) | 8/46 (17) | 0/7 (0) |
| negative (%) | ||||||||||
| Targeted therapy/BRAF | 4/36 (11) | 5/35 (14) | 4/59 (7) | 1/31 (3) | 5/26 (19) | 1/17 (6) | 3/12 (25) | 3/7 (43) | 2/14 (14) | 1/2 (50) |
| untested (%) | ||||||||||
| Chemotherapy/all patients (%) | 26/39 (67) | 27/38 (71) | 52/75 (69) | 45/66 (68) | 60/86 (70) | 54/121 (45) | 60/155 (39) | 51/130 (39) | 27/96 (28) | 3/17 (18) |
| Chemotherapy/BRAF | 2/2 (100) | 1/1 (100) | 6/8 (75) | 17/22 (77) | 19/30 (63) | 18/49 (37) | 24/71 (34) | 23/70 (33) | 7/36 (19) | 1/8 (13) |
| positive (%) | ||||||||||
| Chemotherapy/BRAF | 0/1 (0) | 1/2 (50) | 8/8 (100) | 9/13 (69) | 27/30 (90) | 30/55 (55) | 32/72 (44) | 26/53 (49) | 16/46 (35) | 2/7 (29) |
| negative (%) | ||||||||||
| Chemotherapy/BRAF | 24/36 (67) | 25/35 (71) | 38/59 (64) | 19/31 (61) | 14/26 (54) | 6/17 (35) | 4/12 (33) | 2/7 (29) | 4/14 (29) | 0/2 (0) |
| untested (%) | ||||||||||
Of all patients, 535 (65%) received immunotherapy either before or after brain metastasis, which included interferon (n= 154), granulocyte- macrophage colony-stimulating factor (n=16), interleukin 2 (n=126), ipilimumab (n = 446), PD-1 inhibitors (n=100), or other (n=18); 277 (34%) received targeted therapy, including dabrafenib (n = 94), vemurafenib (n= 137), other BRAF inhibitors (n = 8), trametinib (n = 74), other MEK inhibitors (n = 18), imatinib (n = 3), other C-KIT inhibitors (n = 9), or investigational (n = 25); and 405 (49%) received chemotherapy, including carboplatin (n = 228), paclitaxel (n = 99), temozolomide (n = 262), bevacizumab (n = 29), investigational (n=18), or other (n=105).
Cause of death
The cause of death was known in 485 of the 649 patients (75%) who died during the follow-up period, and the cause was nonneurologic in 304 of 485 (63%).
Comparison to historical cohort
In this series of 823 melanoma patients with brain metastases, not only did BRAF-positive patients survive longer than BRAF-negative patients from the time of initial treatment of the brain metastasis, but the overall survival of 10 months was significantly longer than in our prior report (1985–2005) (10 months vs 6.7 months, P<.01) (18). Furthermore, the data were analyzed before and after ipilimumab and vemurafenib were approved by the Food and Drug Administration (August 2011). There was no significant difference in median survival between patients who received diagnoses of brain metastases from January 2006 to July 2011 and those who received diagnoses from August 2011 to December 2015 (Table 5).
Table 5.
Survival in current and historical cohorts
| Survival from primary diagnosis | Survival from start of BM treatment | |||||
|---|---|---|---|---|---|---|
| Era | MS (IQR), mo | n | P value | MS (IQR), mo | n | P value |
| 1985–2005 | 44.0 (20–88) | 428 | <.01 | 6.7 (3–14) | 481 | <.01 |
| January 2006-July 2011 | 54.7 (29–102) | 363 | Ref | 10.2 (4–28) | 363 | Ref |
| August 2011-Dcccmber 2015 | 55.0 (28–106) | 460 | .23 | 9.6 (4–25) | 460 | .48 |
Abbreviations: BM = brain metastasis; IQR = interquartile range; MS = median survival (Kaplan-Meier estimate); Ref = reference category.
P values compare overall survival of each era versus January 2006 to July 2011.
Discussion
BRAF-positive melanoma patients with brain metastases survive longer than BRAF-negative patients. For individual patients and their physicians, such information about prognosis helps them make choices regarding whether treatment is appropriate and, if so, which treatment is appropriate.
In the larger context, there may be other implications. The management of patients with brain metastases is evolving away from the use of WBRT because of concern about neurocognitive toxicity (22–25). SRS alone is now a standard of care for patients with an increasing number of brain metastases. With the advent of targeted therapies and immunotherapies, an increasing percentage of patients in whom brain metastases develop will be treated with these agents before, after, or both before and after the diagnosis of brain metastases. There is conflicting literature showing both the risk (26, 27) and the reward (28–35) of combining targeted therapies and immunotherapies with SRS.
The risk of such treatment includes pseudoprogression, cerebral edema, and delayed vasculitic leukoencephalopathy with T-cell infiltration (pathologically confirmed in patients who required craniotomy after anti—PD-1 immunotherapy and SRS) (26). Another study showed an increased risk of symptomatic radiation necrosis with SRS and BRAF inhibitors (11% for SRS alone and 28% for SRS combined with BRAF inhibition) (27).
The underlying mechanism of improved survival in this and other retrospective studies remains unclear, but multiple studies have suggested that the combination of SRS with targeted therapies or with immunotherapy and/or targeted therapies not only is well tolerated (31–34) but may yield a survival benefit (28, 35). Retrospective analysis of data from 2 prospective anti—PD-1 (nivolumab) trials showed that the combination of nivolumab and SRS was well tolerated and that local control of brain metastases and overall survival appeared improved compared with historical controls (28).
What, if anything, is appropriate to conclude from conflicting literature? The most apparent difference in the management between the 2 treatment eras is the use of targeted therapies and immunotherapy. SRS alone or in combination with WBRT has improved 1-year local control rates of brain metastases to approximately 80% to 85% (22–25), but use of SRS was similar in both eras. There are limited data on immunotherapy alone for brain metastases. In a 2-arm prospective phase 2 trial of ipilimumab used as a single agent, the responses rates were 25% and 10% in melanoma patients with brain metastases who were asymptomatic not requiring steroids and symptomatic requiring steroids, respectively (36). A preliminary report of a phase 2 trial of pembrolizumab alone in patients with brain metastases showed a response was seen in 4 of 18 patients (22%) with melanoma (37, 38).
So, if immunotherapy alone offers limited response rates for melanoma patients with brain metastases and combined SRS and targeted therapies and/or immunotherapies offer greater response rates and improved survival in this and other retrospective series, it is reasonable to argue that the combination is additive and synergistic and that the drugs act as radiosensitizers or, conversely, the radiation induces an enhanced immune response (abscopal effect). These data are consistent with the mounting literature (11–16, 29, 30) suggesting radiation may induce such an effect in this patient population, but these data cannot distinguish which, if any, of these possible mechanisms are responsible for the findings. Furthermore, there are other possible explanations for the apparent improvement in survival: lead-time bias from improved and/or more frequent imaging resulting in earlier disease detection and treatment, as well as selection bias inherent in any retrospective study.
Another concern that may result in missed therapeutic opportunity is the recent discovery of discordance in BRAF status between the primary tumor and brain metastases (39, 40). This finding suggests BRAF status should be assessed for both the primary and brain metastases whenever clinically feasible.
The observation that extracranial response to BRAF inhibitors exceeds intracranial response led to investigation of the relative concentration of vemurafenib in plasma and cerebrospinal fluid. The mean concentrations of vemurafenib in plasma and cerebrospinal fluid were 53.4 mg/L and 0.5 mg/L (approximately 1%), respectively (41). Reasonable explanations for the seemingly contradictory findings that BRAF inhibitors do not cross the blood-brain barrier and yet BRAF-positive patients with brain metastases survive longer than BRAF-negative patients with brain metastases include that: (1) BRAF-positive melanoma is an inherently less biologically aggressive disease, and survival after the diagnosis of brain metastases is unrelated to the use of targeted therapies; and/or (2) the BRAF inhibitors may enhance the radiosensitivity of BRAF-positive melanoma brain metastases, but BRAF inhibitors alone are inadequate treatment in these patients. This conclusion is consistent with a recent review and other literature that suggested an enhanced response in some series but no response in others and have yet to show that BRAF inhibitors extend survival of melanoma patients with brain métastasés (42, 43).
Many other questions remain, such as how the risk and reward of multimodality therapy varies by gene mutation; other prognostic factors; and/or radiation dose, volume, and fractionation. Prospective trials are needed to answer these questions and better define the effect of gene mutations, BRAF and/or MEK targeted drugs, and immunotherapy in melanoma patients with brain metastases. Until the results of these and other studies are known, retrospective data such as those presented in this article provide imperfect insights but nonetheless illuminate prognosis, as well as current practice patterns, and are hypothesis generating.
Several limitations must be noted: (1) The database used is retrospective with inherent selection bias; (2) the myriad types, combinations, sequences, and timing of the targeted therapies and immunotherapies, as well as the type and dose of radiation therapy, preclude any conclusion from these data regarding which treatment is most effective in this patient population; and (3) limited data were available regarding the toxicity of combined-modality therapy. Last, transition to a diagnosis-specific GPA that incorporates molecular variables (molecular melanoma GPA), such as gene status, is needed to more accurately estimate survival for these patients in the modem era of targeted therapies and immunotherapy, guide clinical decision making, and stratify future clinical trials to ensure comparison of similar patient groups.
Acknowledgments—
The authors acknowledge the database support and management provided by Susan Lowry, Database Programmer/Analyst and REDCap Administrator, Biostatistical Design and Analysis Center, Clinical and Translational Science Institute, University of Minnesota, Minneapolis, Minnesota.
This work was presented at the American Society for Radiation Oncology Annual Meeting; September 25, 2016; Boston, MA.
Funding in the form of grant support was provided by the following: (1) National Institutes of Health (NIH) grant number UL1TR000114 from the National Center for Advancing Translational Sciences. Study data were collected and managed using REDCap electronic data capture tools hosted at the University of Minnesota; (2) NIH grant number P30 CA77598 using the Biostatistics and Bioinformatics Core shared resource of the Masonic Cancer Center, University of Minnesota, and National Center for Advancing Translational Sciences. The design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication were solely the responsibility of the authors, and the article does not necessarily represent the official views of the funders or sponsors (National Center For Research Resources or NIH).
The authors report relationships with the following: Varian (J.P.K.); Genentech (H.A.S.); Varian, Siemens, Accuray, BrainLab, and Elekta (D.R.); and Abbott, Novelos, Phillips, BMS, Celldex, Roche, Elekta, Novocure, Novartis, Cavion, and Pharmacyclics (M.M.).
Footnotes
Conflict of interest:
The other authors report no conflict of interest.
References
- 1.Hodi FS, O’Day SJ, McDermott DF, et al. Improved survival with ipilimumab in patients with metastatic melanoma [published erratum appears in N Engl J Med 2010;363:1290]. N Engl J Med 2010;363:711–723. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Robert C, Schachter J, Long GV, et al. Pembrolizumab versus ipilimumab in advanced melanoma. N Engl J Med 2015;372:2521–2532. [DOI] [PubMed] [Google Scholar]
- 3.Postow MA, Chesney J, Pavlick AC, et al. Nivolumab and ipilimumab versus ipilimumab in untreated melanoma. N Engl J Med 2015;372: 2006–2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chapman PB, Hauschild A, Robert C, et al. Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N Engl J Med 2011;364:2507–2516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Flaherty KT, Robert C, Hersey P, et al. Improved survival with MEK inhibition in BRAF-mutated melanoma. N Engl J Med 2012;367:107–114. [DOI] [PubMed] [Google Scholar]
- 6.Flaherty KT, Infante JR, Daud A, et al. Combined BRAF and MEK inhibition in melanoma with BRAF V600 mutations. N Engl J Med 2012;367:1694–1703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin 2016;66:7–30. [DOI] [PubMed] [Google Scholar]
- 8.Davies MA, Liu P, McIntyre S, et al. Prognostic factors for survival in melanoma patients with brain metastases. Cancer 2011; 117:1687–1696. [DOI] [PubMed] [Google Scholar]
- 9.Kanner AA, Barnett GH. Management of brain metastases in malignant melanoma patients In: Sawaya R, editor. Intracranial Metastases: Current Management Strategies. Malden, MA: Futura, Blackwell Publishing; 2004. p. 245–265. [Google Scholar]
- 10.Tas F Metastatic behavior in melanoma: Timing, pattern, survival, and influencing factors. J Oncol 2012;2012:647684. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Demaria S, Ng B, Devitt MI, et al. Ionizing radiation inhibition of disease untreated tumors (abscopal effect) is immune mediated. Int J Radial Oncol Biol Phys 2004;58:862–870. [DOI] [PubMed] [Google Scholar]
- 12.Burnette B, Weichselbaum RR. Radiation as an immune modulator. Semin Radial Oncol 2013;23:273–280. [DOI] [PubMed] [Google Scholar]
- 13.Ahmed MM, Hodge JW, Guha C, et al. Harnessing the potential of radiation-induced immune modulation for cancer therapy. Cancer Immunol Res 2013;1:280–284. [DOI] [PubMed] [Google Scholar]
- 14.Hiniker SM, Chen DS, Knox SJ. Abscopal effect in a patient with melanoma. N Engl J Med 2012;366:2035–2036. [DOI] [PubMed] [Google Scholar]
- 15.Postow MA, Callahan MK, Barker CA, et al. Immunologic correlates of the abscopal effect in a patient with melanoma. N Engl J Med 2012; 366:925–931. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Stammell EF, Wolchok JD, Gnjatic S, et al. The abscopal effect associated with a systemic anti-melanoma immune response. Int J Radial Oncol Biol Phys 2013;85:293–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Suki D The epidemiology of brain metastases In: Sawaya R, editor. Intracranial Metastases: Current Management Strategies. Malden, MA: Futura, Blackwell Publishing; 2004. p. 20–34. [Google Scholar]
- 18.Sperduto PW, Chao ST, Sneed PK, et al. Diagnosis-specific prognostic factors, indexes, and treatment outcomes for patients with newly diagnosed brain metastases: A multi-institutional analysis of 4,259 patients. Int J Radiat Oncol Biol Phys 2010;77:655–661. [DOI] [PubMed] [Google Scholar]
- 19.Sperduto PW, Kased N, Roberge D, et al. Summary report on the Graded Prognostic Assessment (GPA): An accurate and facile diagnosis-specific tool to estimate survival, guide treatment and stratify clinical trials for patients with brain metastases. J Clin Oncol 2012;30:419–425. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Sperduto PW, Yang TJ, Beal K, et al. The effect of gene alterations and tyrosine kinase inhibition on survival and cause of death in patients with adenocarcinoma of the lung and brain metastases. Int J Radiat Oncol Biol Phys 2016;96:406–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sperduto PW, Yang TK, Beal K, et al. Improved survival and prognostic ability in lung cancer patients with brain metastases: An update of the Graded Prognostic Assessment for lung cancer using molecular markers (Lung-molGPA) le-pub ahead of print]. JAMA Oncol 2016; 10.1001/jamaoncol.2016.3834. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Aoyama H, Shirato H, Tago M, et al. Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic radiosurgery alone for treatment of brain metastases: A randomized controlled trial. JAMA 2006;295:2483–2491. [DOI] [PubMed] [Google Scholar]
- 23.Kocher M, Soffietti R, Abacioglu U, et al. Adjuvant whole-brain radiotherapy versus observation after radiosurgery or surgical resection of one to three cerebral metastases: Results of the EORTC 22952– 26001 study. J Clin Oncol 2010;29:134–141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Chang EL, Wefel JS, Hess KR, et al. Neurocognition in patients with brain metastases treated with radiosurgery or radiosurgery plus whole-brain radiation: A randomized controlled trial. Lancet Oncol 2009; 10: 1037–1044. [DOI] [PubMed] [Google Scholar]
- 25.Brown PD, Jaeckle K, Ballman KV, et al. Effect of radiosurgery alone vs radiosurgery with whole brain radiation therapy on cognitive function in patients with 1 to 3 brain metastases: A randomized clinical trial. JAMA 2016;316:401–409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Alomari AK, Cohen J, Vortmeyer AO, et al. Possible interactions of anti-PD-1 therapy with the effects of radiosurgery on brain metastases. Cancer Immunol Res 2016;4:481–487. [DOI] [PubMed] [Google Scholar]
- 27.Chowdhary M, Patel K, Switchenko J, et al. Increased risk of radiation necrosis following BRAF inhibitor therapy and stereotactic radiosurgery (SRS). Int J Radiat Oncol Biol Phys 2016;96(Suppl.):S59. [Google Scholar]
- 28.Ahmed KA, Stallworth DG, Kim Y, et al. Clinical outcomes of melanoma brain metastases treated with stereotactic radiation and anti- PD-1 therapy. Ann Oncol 2016;27:434–441. [DOI] [PubMed] [Google Scholar]
- 29.Sharabi AB, Nirschi CJ, Kochel CM, et al. Stereotactic radiation therapy augments antigen-specific PD-1-mediated anti-tumor responses via cross-presentation of tumor antigen. Cancer Immunol Res 2015;3:345–355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Sharabi AB, Tran PT, Lim M, et al. Stereotactic radiotherapy combined with immunotherapy: Augmenting radiation’s role in local and systemic treatment. Oncology 2015;29:331–340. [PMC free article] [PubMed] [Google Scholar]
- 31.Knisely JP, Yu JB, Flanigan J, et al. Radiosurgery for melanoma brain metastases in the ipilimumab era and the possibility of longer survival. J Neurosurg 2012;117:227–233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kiess AP, Wolchok JD, Barker CA, et al. Stereotactic radiosurgery for melanoma brain metastases in patients receiving ipilimumab: Safety profile and efficacy of combined treatment. Int J Radiat Oncol Biol Phys 2015;92:368–375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Silk AW, Bassetti MF, West BT, et al. Ipilimumab and radiation therapy for melanoma brain metastases. Cancer Med 2013;2:899–906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Patel KR, Shoukat S, Oliver DE, et al. Ipilimumab and stereotactic radiosurgery versus stereotactic radiosurgery alone for newly diagnosed melanoma brain metastases [e-pub ahead of print]. Am J Clin Oncol 2015; 10.1097/COC.0000000000000199. [DOI] [PubMed] [Google Scholar]
- 35.Wattson DA, Sullivan RJ, Niemierko A, et al. Survival patterns following brain metastases for patients with melanoma in the MAP- kinase inhibitor era. J Neurooncol 2015;123:75–84. [DOI] [PubMed] [Google Scholar]
- 36.Margolin K, Emstoff MS, Hamid O, et al. Ipilimumab in patients with melanoma and brain metastases: An open-label phase 2 trial. Lancet Oncol 2012;13:59–65. [DOI] [PubMed] [Google Scholar]
- 37.Goldberg SB, Gettinger SN, Mahajan A, et al. Pembrolizumab for patients with melanoma or non-small-cell lung cancer and untreated brain metastases: Early analysis of a non-randomised, open-label, phase 2 trial. Lancet Oncol 2016;17:976–983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Cohen JV, Kluger HM. Systemic immunotherapy for the treatment of brain métastasés. Front Oncol 2016;6:49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Saroufim M, Habib RH, Gerges R, et al. Comparing BRAF mutation status in matched primary and metastatic cutaneous melanomas: Implications on optimized targeted therapy. Exp Mol Pathol 2014;97: 315–320. [DOI] [PubMed] [Google Scholar]
- 40.Brastianos PK, Carter SL, Santagata S, et al. Genomic characterization of brain metastases reveals branched evolution and potential therapeutic targets. Cancer Discov 2015;5:1164–1177. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Sakji-Dupré L, Le Rhun E, Templier C, et al. Cerebrospinal fluid concentrations of vemurafenib in patients treated for brain metastatic BRAF-V600 mutated melanoma. Melanoma Res 2015;25: 302–305. [DOI] [PubMed] [Google Scholar]
- 42.Chowdhary M, Patel KR, Danish HH, et al. BRAF inhibitors and radiotherapy for melanoma brain metastases: Potential advantages and disadvantages of combination therapy. Onco Targets Ther 2016;9: 7149–7159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Rochet NM, Dronca RS, Kottschade LA, et al. Melanoma brain metastases and vemurafenib: Need for further investigation. Mayo Clin Proc 2012;87:976–981. [DOI] [PMC free article] [PubMed] [Google Scholar]