Nature Medicine: MSKCC Releases Blockbuster Results of Liquid Biopsy for Lung Cancer! Large international cohorts confirmed that ctDNA is effective in predicting non-trivial prognosis and guiding targeted therapy

In recent years, the treatment and prognosis of patients with non-small cell lung cancer (NSCLC) have greatly improved thanks to targeted therapy, which is mainly based on the wide application of tissue sequencing, but in the real world, there are many patients who cannot perform tissue sequencing for various reasons [1-3], so it is easier to obtain samples and detect blood circulating tumor DNA (ctDNA) "homeopathic"
.

However, there are some uncertainties in young ctDNA, for example, ctDNA-matched targeted therapies have similar response rates to tissue sequencing matches, but studies are mainly conducted in small cohorts and there is not enough follow-up time to determine whether it translates into the same overall survival benefit, and ctDNA can reflect tumor aggressiveness and disease burden, but is its predictive prognostic value independent of tumor size or other clinical factors? By solving these problems, ctDNA can "take it to the next level"
.

Today, the journal Nature Medicine published the results of the study [4], in which researchers at Memorial Sloan Kettering Cancer Center (MSKCC) verified in a large prospective NSCLC international cohort that ctDNA positivity was associated with lower overall survival (OS) (HR=2.
05, p<0.
001), and that patients who received matched targeted therapy after ctDNA sequencing had higher OS than those who did not receive (HR= 0.
63，p＜0.
001）<b10>。

The researchers identified ctDNA genomic variants undetected by tissue sequencing in 1/4 of patients, which characterize a number of genetic variants that drive drug resistance, including the RIC TOR and PIK3CA variants
.

This study shows that ctDNA not only guides treatment decisions, but also has prognostic value
in itself.

The study included a total of 1127 patients with stage IV or recurrent NSCLC who either had no known driver mutations prior to enrollment or had disease progression
after targeted therapy.

All patients underwent ctDNA sequencing, blood samples were obtained on the first day of enrollment, and patients had the option to provide blood samples again later, although the majority (89%) provided only one blood sample
.
All patients enrolled in MSKCC had tissue sequencing
.
Patients can undergo additional plasma circulating cell-free DNA (cfDNA) sequencing
if they wish.

ctDNA sequencing has a lower failure rate and a shorter
turnaround time compared to tissue.
164 (13%) of the 1219 tissue sequences failed due to insufficient material or sequencing failure, compared to only 37 (2%) of the 1919 plasma sequencing (ctDNA or cfDNA), and 2 patients had unsuccessful plasma and tissue sequencing
.
From sample collection to sequencing report, tissue sequencing took 33 days, while plasma sequencing took only 11 days
.

64% (722/1127) of patients had at least one ctDNA variant, and those with ctDNA variants had lower OS (HR = 2.
05, p<0.
001), which can be generalized to treatment-naïve and treated patients in the MSKCC and Sydney cohorts, and the higher the number of ctDNA variants, the lower<b10> the OS.
Some patients had only variants with unknown effects, but they also had lower OS than those without variation, suggesting that a positive ctDNA result predicted a worse prognosis even without pathogenic variation
.

OS of patients with ctDNA variant (blue) and no variation

Tumor mutational burden (TMB) is not associated with ctDNA, which is an independent predictor
of worse prognosis in patients receiving targeted therapy, immunotherapy, or chemotherapy.
Therefore, the predictive role of ctDNA is independent of the treatment model and tumor genomics
.

In addition to these two, multiple multivariate analyses confirmed that ctDNA has independent prognostic value
independent of multiple factors such as tumor metabolic volume.

After sequencing results, a total of 418 (37%) patients received targeted therapy, of which 255 (23%) were ctDNA-matched and 163 (14%) were matched by tissue sequencing
.
In all patients receiving targeted therapy, the OS of ctDNA was lower, and the OS of patients who received ctDNA matching to targeted therapy was higher (HR=0.
63, p<0.
001) compared with patients who did not receive targeted therapy, while whether or not ctDNA was not detected did not affect OS<b11>.

This suggests that targeted therapy has a greater relative benefit
in ctDNA-positive patients.

Of the 429 patients who received tissue sequencing within 30 days of ctDNA sequencing, 109 (25%) only had mutations or copy number changes detected in ctDNA sequencing, of which 3 had EGFR T790M mutations, so the patients switched therapy, and other targeted mutations included 1 KRAS G12C mutation, 1 BRAF V600E mutation, and 1 RET fusion, but no switching therapy
.

Patients with only the variation detected by ctDNA had a reduced OS compared to those with variants that could be detected by both ctDNA and tissue sequencing, and the results were consistent
in both mutation and copy number changes.
This suggests that variants that can be detected only by ctDNA are also relatively common and equally valuable in predicting prognosis
.

Only patients with ctDNA detected variants (violet) and patients with both ctDNA and tissue sequencing detected the OS of patients with mutations (blue).

The ctDNA-only variants the researchers found were also common in tissues, with a high proportion of secondary resistance genes, particularly RICTOR, NTRK1, MET and ERBB2 amplification and PIK3CA mutations
.

Only ctDNA and tissue variants, pathogenic variants are distributed at the gene level

These secondary resistance gene variants are generally associated with a worse prognosis, but the analysis showed that patients with ALK, RET and EGFR variants detected had higher OS, patients with STK11 variants had lower OS, and RICTOR, NTRK1, MET, and ERBB2 and PIK3CA variants were not associated with
lower OS.
This suggests that differences in ctDNA-tissue sequencing results are a key prognostic factor rather than genomic variation
.

Overall, this study shows that ctDNA can guide targeted therapy in NSCLC patients on the one hand, improve patient outcomes, and on the other hand, it can also be used as an independent OS predictor, which is a marker of tumor micrometastasis or increased tumor aggressiveness, and can be combined
with radiological test results when assessing disease burden and recurrence risk after systemic treatment.

Of course, these results also raise new questions for researchers, such as, why do ctDNA-negative patients have higher OS? Is it because their tumors proliferate slowly? Is it related to specific features of the tumor microenvironment that affect drug uptake? Can we turn ctDNA positive into negative with drugs? These questions need to be based on ctDNA dynamics, starting with tumors and patients, to elucidate the clinical and biological mechanisms [5].

References:

[1] Bruno D S, Hess L M, Li X, et al.
Racial disparities in biomarker testing and clinical trial enrollment in non-small cell lung cancer (NSCLC)[J].
J Clin Oncol, 2021.

[2] Robert N J, Nwokeji E D, Espirito J L, et al.
Biomarker tissue journey among patients (pts) with untreated metastatic non-small cell lung cancer (metastatic NSCLC) in the US Oncology Network community practices[J].
J Clin Oncol, 2021, 39.

[3] Zugazagoitia J, Ramos I, Trigo J M, et al.
Clinical utility of plasma-based digital next-generation sequencing in patients with advance-stage lung adenocarcinomas with insufficient tumor samples for tissue genotyping[J].
Annals of Oncology, 2019, 30(2): 290-296.

[4]

[5]