Frontiers of ctDNA, Targeted Therapies, and Immunotherapy in Non-Small-Cell Lung Cancer

Chennianci Zhu; Weihao Zhuang; Limin Chen; Wenyu Yang; Wen-Bin Ou


Transl Lung Cancer Res. 2020;9(1):111-138. 

In This Article

Significant Links Between ctDNA, TKI Targeted Therapy, and Immunotherapy

ctDNAs as Prognostic Markers to Determine Acquired Resistance and the Effectiveness of TKIs

Even when patients exhibit an initial disease progression event observed by the monitoring of ctDNA levels, continued TKI treatment [osimertinib,[86] afatinib[222]] can remain effective. In a phase III trial, continued afatinib treatment after disease progression doubled the patient PFS (5.6 months with afatinib treatment vs. 2.8 months with chemotherapy).[222] When a drug continues to suppress the majority of tumor cells, compensatory pathways may be upregulated. Upon drug withdrawal, dramatic disease progression called "tumor flare" might occur.[223] However, treatment should be based on each patient's mutation profile; for instance, a patient with the EGFR T790M mutation might exhibit disease progression after rociletinib treatment for a certain duration.[7] In this case, a change in EGFR-TKI from rociletinib to osimertinib seems to be a better strategy than continued rociletinib treatment. A prolonged PR to osimertinib after PD with rociletinib treatment was observed in EGFR T790M-positive NSCLC patients.[224,225]

The ctDNA levels in EGFR T790M-positive patients exhibited a specific, similar, and sudden spike that developed at different time points during routine ctDNA profiling after TKI treatment, but this pattern was not observed in EGFR T790M-negative patients.[70] This result illustrates the possibility of predicting cancer progression by analyzing ctDNA profiles and determining the possible mutation type. Analysis of ctDNA shedding in samples taken before and after disease progression showed a positive correlation between the level of ctDNA shedding and tumor burden, suggesting that disease in these non-shedding patients is entirely controlled at progression. The emergence of clinically important resistance may correlate with the presence of detectable ctDNA.[226] Resistance caused by the EGFR T790M mutation has irreversible outcomes but can be detected by early ctDNA analysis, which can identify the mutation even before it has a large impact on patients.[64]

EGFR T790M ctDNA profiling also provides solid evidence of the effectiveness of TKIs and indications for drug switching strategies. EGFR T790M was strongly correlated with large increases in mutant DNA levels. This strongly suggests switching from first-generation TKIs, which no longer effectively control disease progression, to third-generation TKIs with the need of intervention at an earlier time.[67] Therefore, EGFR T790M has emerged as a favorable prognostic marker to determine acquired resistance and an important predictive marker for the effectiveness of TKIs.

ctDNAs as Predictive Markers for Immunotherapy to Assess Pseudoprogression and Blood TMB (bTMB)

ctDNAs are widely used as an efficient clinical tool to monitor the effects of chemotherapy and targeted therapy. Recently, more and more studies focused on the application of ctDNA assays in monitoring responses of immunotherapy by ICIs.

ctDNA levels correlated with tumor progression in patients receiving anti-CTLA-4 and anti-PD-L1 therapy, and pseudoprogression was revealed by undetectable ctDNA 3 weeks prior to clinical improvement.[227] Other studies[228–230] also proved the efficiency of ctDNA, decreased ctDNA levels were observed in response to treatment several weeks after initiation. Guibert et al.[231] monitored responses to anti-PD-1 treatment of KRAS mutated lung adenocarcinoma by ddPCR on plasma ctDNA and discriminated pseudo from true progression. Patients with pseudoprogression often began with undetectable ctDNA at baseline or detectable ctDNA at baseline followed by a greater than 10-fold decrease in ctDNA level in response to treatment;[232] these patients with low or undetected ctDNA level at the beginning of and during therapy often have better responses to immunotherapy.[233] Patients with more than 50% decrease compared to ctDNA baseline levels showed superior PFS and OS than those by less than 50%.[234] Baseline ctDNA level, and ctDNA level post-treatment, could be prognostic factors in patients receiving ICIs.

ctDNA has been investigated as a predictive marker that reveals TMB in response to immunotherapy.[235–237] TMB assessed by targeted NGS was significantly associated with improved benefit among patients with NSCLC treated with ICIs.[238] Recent studies[239–241] investigated bTMB profiled with ctDNA sequencing and whether it predicts responses to immunotherapy. bTMB quantified by NGS were highly concordant with tissue TMB (tTMB), so that TMB can be accurately measured in plasma. bTMB also associated with clinical benefit such as longer PFS in patients treated with atezolizumab.[240] Although there are only a few relative studies, researchers positively concluded that bTMB quantified by ctDNA had the potential to be a novel biomarker for prognosis in patients with NSCLC treated by immunotherapy. However, due to the limitation of sequencing depth, in order to quantify TMB by ctDNA, mutations with a minimum allele frequency must be above 1% in ctDNA for a bTMB score to be valid.[240]

Combining TKI Therapy and Immunotherapy

Many studies have pointed out, EGFR/ALK status in NSCLC cell lines were correlated with the expression level of PD-(L)1. For example, EGFR activating mutations, EGFR T790M, EML4-ALK fusion, and MET overexpression were significantly associated with increased expression of PD-L1.[192,193] However, the number of studies investigating how TKI treatment affects immunotherapy is still limited.

Although EGFR mutations or ALK rearrangements were associated with low ORRs to PD-1/PD-L1 inhibitors,[242] and EGFR T790M upregulated PD-L1 level;[243] It is contradictory to our current understanding that Haratani et al.[244] found out that 25 patients with EGFR mutation but negative for T790M benefited more from nivolumab treatment after EGFR-TKI therapy, possibly because of higher PD-L1 expression levels in those T790M-negative patients. A few nivolumab responders with high levels of PD-L1 expression also experienced copy number gain for MET. Prospective clinical trials are required to confirm the efficacy of anti-PD-1 treatment for EGFR T790M-negative patients. Another recent study[245] focused on combining erlotinib/gefitinib with pembrolizumab as first-line therapy for NSCLC patients with EGFR sensitizing mutations. ORRs were 41.7% and 14.3% in pembrolizumab plus erlotinib and pembrolizumab plus gefitinib, respectively. However, there were no improvement of ORRs compared to previous monotherapies, and 5 in 7 patients receiving pembrolizumab plus gefitinib were observed to have drug-related liver toxicity. Patients received pembrolizumab plus erlotinib were observed to have adverse events similar to monotherapy, demonstrating the safety profile of this combination. Combining immunotherapy with EGFR-TKI therapy might be reasonable, nevertheless, due to the small size of cohort in this study, further evaluation is still required. In another clinical trial (NCT02013219),[246] atezolizumab plus erlotinib demonstrated a manageable safety profile and achieved ORR of 75%, a great improvement from previous atezolizumab monotherapy. Trial (NCT02088112)[247] on durvalumab plus gefitinib achieved ORR of 78.9%. Trials ongoing (NCT02013219, NCT02088112, NCT02364609, and NCT02143466) all involve the combination of EGFR/ALK-TKI therapy with immunotherapy, a novel treatment strategy that is still at an early stage of exploration.