Introduction: A BRAFV600E mutation is an unfavorable prognostic biomarker for CRC and is associated with short-lived treatment response to BRAF and EGFR blockade. Anti-PD-1 therapies are ineffective in MSS CRC but demonstrate efficacy in combination with BRAF + EGFR inhibition for MSS BRAFV600E CRC. A comprehensive characterization of MSS BRAFV600E CRC as an immunologically distinct subpopulation of MSS CRC has not been performed. Here, we characterize the clinicopathological and transcriptomic features of MSS BRAFV600E CRC patients, relative to BRAFWT MSS CRC patients.
Methods: De-identified records of 12,009 MSS CRC patients were retrospectively analyzed from the Tempus clinicogenomic database. CMS subtypes were derived using the CMScaller algorithm. Survival was assessed using Kaplan-Meier survival analysis with risk set adjustment (RSA) method; log-rank test was used to compare survival. Categorical and continuous variables were compared using chi-squared test and Wilcoxon rank sum test, respectively.
Results: BRAFV600E was found in 5% (n=630) of the patients. Demographics and clinicopathological features varied significantly between the BRAFWT and BRAFV600E cohorts (Table 1). Relative to the BRAFWT cohort, BRAFV600E tumors were enriched for the immune-activated CMS1 signature (53% vs 11%, p< .001), PD-L1 detection (9.5% vs 3%, p<0.001), and CD8+ T-cell infiltration (4.6% vs. 2.5%, p<0.001). CMS4 was linked to an unfavorable prognosis in the BRAFWT cohort (HR=0.58 compared to CMS1, p=0.026), yet this was not observed in the BRAFV600E cohort (HR=2.60 compared to CMS1, p=0.083).
Conclusions: BRAFV600E MSS CRC exhibited immune activation characteristics that were not observed in the BRAFWT group, including greater CMS1 status, PD-L1 expression, and CD8-T cell infiltration. Our findings support investigation of novel immune-based therapeutic strategies of MSS BRAFV600E CRC as an immunologically distinct subpopulation of MSS CRC.
Cohort characteristics
|
|
|
BRAFWT |
BRAFV600E |
P-value |
Demographics |
Age (median) |
|
60 (51,69) |
65 (55,74) |
<0.001 |
|
Ethnicity |
Not Hispanic or Latino |
3,444 (83%) |
200 (93%) |
0.001 |
|
|
Hispanic or Latino |
699 (17%) |
16 (7.4%) |
|
|
Gender |
Female |
4,771 (42%) |
345 (55%) |
<0.001 |
|
|
Male |
6,608 (58%) |
285 (45%) |
|
|
Race |
White |
5,050 (75%) |
318 (87%) |
<0.001 |
|
|
Black or African American |
888 (13%) |
19 (5.2%) |
|
|
|
Other |
517 (7.7%) |
20 (5.4%) |
|
|
|
Asian |
297 (4.4%) |
10 (2.7%) |
|
Clinicopathology |
TNM stage |
I |
61 (0.7%) |
0 (0%) |
0.1 |
|
|
II |
388 (4.6%) |
17 (3.2%) |
|
|
|
III |
1,130 (13%) |
88 (17%) |
|
|
|
IV |
6,941 (81%) |
424 (80%) |
|
|
Laterality |
Right |
777 (15%) |
89 (39%) |
<0.001 |
|
|
Transverse |
193 (3.8%) |
27 (12%) |
|
|
|
Left |
2,326 (46%) |
73 (32%) |
|
|
|
Rectum |
1,761 (35%) |
37 (16%) |
|
|
CMS |
CMS1 |
1,077 (11%) |
307 (53%) |
<0.001 |
|
|
CMS2 |
3,067 (31%) |
6 (1%) |
|
|
|
CMS3 |
1,777 (18%) |
71 (12%) |
|
|
|
CMS4 |
3,920 (40%) |
199 (34%) |
|
Immunology |
CD8 T-cell infiltration (median, IQR) |
|
2.5% (0.0, 7.7) |
4.6% (0.0, 9.4) |
<0.001 |
|
PD-L1 |
|
127 (3%) |
22 (9.5%) |
VIEW THE PUBLICATION
VIEW THE POSTER