Abstract 363P
Background
While several studies have explored the evolving clonal profile of mCRC under stress of EGFRi using circulating tumor DNA (ctDNA), the heterogeneity and small samples limit our understanding of true prevalence of acq-GAs associated with EGFR resistance, and potential for concurrent chemotherapy to alter resistance patterns.
Methods
A systematic literature search (PubMed + Google Scholar) was performed for studies assessing acq-GAs in ctDNA post progression on EGFRi based therapy (± chemotherapy) in mCRC. Data was extracted from 13 identified studies, categorized by their use of EGFRi in either first-line (1L) (N = 2) or second-line and beyond (2L+) (N = 11). The primary objective was to assess prevalence of acq-GAs of interest, expressly, those established as resistance mechanisms to EGFRi: KRAS, NRAS, BRAF, EGFR ectodomain mutations and ERBB2 and MET amplification. The pooled prevalence was calculated using a fixed effect model (heterogeneity assessed by I2 statistic).
Results
Across 13 examined studies, the net sample size was 825 (range 4 - 331) [1L (138) and 2L+ (687)] and 325 patients acq GAs at progression on EGFRi with a pooled prevalence of 39% (95%CI 36 – 41). This varied substantially by line of therapy with prevalence of 8% (95%CI 3 – 12) in 1L studies (I2 0%) significantly lower than 55% (95%CI 52 – 58) seen in 2L+ setting (I2 95%) (OR 0.07, 95%CI 0.04 – 0.13, P < 0.0001). Prevalence of specific acq-GAs was also lower in 1L compared to 2L+ studies (Table). Table: 363P
| Genomic Alteration | Pooled Prevalence1 | Comparison2 | ||
| 1L (%) | 2L+ (%) | OR, 95%CI | P | |
| Mutations | ||||
| All RAS | 5 | 40 | 0.08, 0.03-0.17 | <0.0001 |
| KRAS | 2 | 29 | 0.05, 0.02-0.15 | <0.0001 |
| NRAS | 1 | 6 | 0.12, 0.01-0.66 | 0.0083 |
| BRAF | 1 | 9 | 0.08, 0.00-0.42 | 0.0002 |
| EGFR-ectodomain (ECD) | 2 | 20 | 0.09, 0.03-0.26 | <0.0001 |
| Amplifications | ||||
| ERBB2 | 2 | 7 | 0.23, 0.02-2.10 | 0.25 |
| MET | 0 | 17 | 0.00, 0.00-0.31 | 0.0008 |
| 1. Pooled prevalence and its corresponding 95% confidence intervals (95%CI) were calculated by meta-analysis of extracted proportions of key GAs with Jamovi software. 2. Odds ratio (OR) and 95%CI were calculated using Fisher-exact test for comparison of prevalence between groups. |
Conclusions
Our analysis demonstrates that acquisition of key GAs associated with resistance to EGFRi in mCRC is rare with upfront use of anti-EGFR mAbs compared to later lines, indicating differing resistance mechanisms. This has implications for prospective trials examining EGFRi rechallenge strategies and its clinical use guided by acq-GAs and highlights need to address alternate resistance mechanisms.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
MD Anderson Cancer Center, Houston, TX, USA.
Disclosure
All authors have declared no conflicts of interest.