Prediction of prognosis, monitoring of recurrence, application and breakthrough of ctDNA in the diagnosis and treatment of gastric cancer

*For medical professionals only

Professor Zhang Xiaotian takes you to learn the new progress of liquid biopsy for gastric cancer!

Liquid biopsy refers to the test performed on a blood sample to look for circulating tumor cells or pieces of tumor cell DNA in the blood, and liquid biopsy can be used to help detect cancer early, guide treatment, detect cancer recurrence, etc
.
Compared with tissue biopsy technology, liquid biopsy is relatively non-invasive, and through the detection of blood samples at multiple time points, it can dynamically monitor and dynamically reflect the tumor burden and tumor variation in patients in real time, which has great clinical application value
.
Among the various liquid biopsy techniques, circulating tumor DNA (ctDNA) is one of
the most abundant application scenarios.

On November 9, 2022, at the 25th National Congress of Clinical Oncology and the 2022 CSCO Annual Conference, Professor Zhang Xiaotian from Peking University Cancer Hospital gave a wonderful speech on the topic of "Application and Breakthrough of ctDNA in the Diagnosis and Treatment of Gastric Cancer".

Figure 1 Professor Zhang Xiaotian's academic lecture at the CSCO conference

Guardant360 clinical trial evidence showed that gastric cancer ctDNA NGS and tissue HER2 status (IHC/FISH) were 61% (17/28), sensitivity was 58% (15/26), and specific was 100% (2/2).

Colorectal cancer blood NGS and tissue (FISH or NGS) detected 83% (62/75) of HER2 amplification, PPA of 82% (32/39), and NPA of 83% (30/36).

Professor Zhang Xiaotian introduced that IHC/FISH technology is the gold standard for tissue HER2 detection, but the positive test of HER2 in gastric cancer still faces the dilemma of heterogeneity and tissue accessibility: HER2 molecules have spatiotemporal heterogeneity, the test results of tumor tissue sampling may not reflect the whole tumor picture, and IHC detection only for major lesions may affect the treatment decisions
of some patients.

The use of ctDNA detection can reflect the HER2 status of tumor tissue or even the entire genome state, comprehensively predict and guide the treatment of HER2-positive gastric cancer, replace re-biopsy, and evaluate the efficacy
of drugs.

Clinical evidence showed that ctDNA NGS detection was performed on blood samples of 40 patients with advanced gastric cancer receiving anti-HER2 therapy at baseline, after treatment and at the time of disease progression, and by analyzing the abundance of TP53 and HER2 mutations, 40% of patients could predict disease progression before CT evaluation, and the trend
of disease progression could be detected on average 38.
4 days earlier than CT.

In terms of prognosis detection, NGS detection of CTDNA in 46 patients with advanced gastric cancer treated with PD-1 monoclonal antibody showed a 25% decrease in maximal somatic variant allele frequency (maxVAF) than PFS (7.
3 months vs 3.
6 months) and a higher response rate (53.
3% vs 13.
3%, p=0.
06)
than PFS without decrease 。 The mutational status of TGFBR2, RHOA and PREX2 affected the PFS (p<0.
05) of immunotherapy, and patients with CEBPA, FGFR4, MET or KMT2Bs variants had a high probability of immune-related adverse events (P=0.
09).

Figure 2.
ctDNA gene enrichment plot of gastric cancer patients in different groups

The analysis of ctDNA detection of blood samples of gastric cancer patients receiving immunotherapy combined with anti-HER2 therapy showed that HER2 amplification and POLE variants were significantly enriched in the group
of patients with the best efficacy evaluated as partial response and complete remission.
The "immune-related negative gene set" variants including RTK/RAS, PI3K, p53, cell cycle, WNT, and ICI_neg_genes harmful variants were significantly enriched in patients with stable disease and disease progression
.

Common mutations in POLE are p.
P286R, V411L, S297F, A456P and S459F, indicating a positive correlation
with immune efficacy.
However, the significance of clinical guidance for some uncommon variants is still unclear, and the above studies further propose that Polyphen-2 predicts harmful variants: p.
Y766F, p.
K733N, p.
L1766M and p.
A546S&p.
A2262S
.

Patients with harmful variations of POLE with ctDNA had a higher ORR than
those with ctDNA carrying the POLE wild-type gene.
However, ctDNA carrying "immune-related negative gene set" variation was significantly correlated with poor efficacy of combination therapy, and ctDNA "immune-related negative gene set" variation could be used as an independent predictor
of "immunotherapy + anti-HER2 combination therapy".

In gastric cancer patients undergoing surgery, minimal residual disease detection has limited data for predicting recurrence risk, while ctDNA can effectively predict disease recurrence
in patients with gastroesophageal cancer.

In a clinical study of 269 patients with stage I-IV gastroesophageal cancer (esophageal cancer N=79, gastroesophageal junction carcinoma N=73, gastric cancer N=117), a personalized PCR test was used to detect ctDNA in patients' blood samples, and in the postoperative detection cohort of stage I-III patients, compared with ctDNA-negative patients, The worse prognosis for ctDNA-positive patients (HR-9.
0, p6.
S0e-12) suggests that our ctDNA-positive patients require more aggressive treatment
.

Compared with conventional imaging, ctDNA can detect disease recurrence
about 6 months earlier.
In a clinical study of 46 patients with gastric cancer who underwent surgical resection of stages I-III, the variation detected by ctDNA into tissue was defined as ctDNA-positive, and the detection rate of ctDNA in preoperative blood was 45%.

Among them, the ctDNA detection rate of stage III patients was 68% (15/22), and the ctDNA detection rate of stage I-II patients was 21% (4/9).

Figure 3.
ctDNA testing can detect cancer recurrence early

Compared to imaging, ctDNA can detect recurrence
up to 179 days earlier.
Professor Zhang Xiaotian mentioned that if ctDNA detection technology detects disease recurrence earlier than imaging technology, whether intervention is needed in advance is also a question
that we need to consider.

The PANGEN study analyzed the molecular variation patterns of 80 gastric cancer primary and metastases, and about 35% of gastric cancer patients (28/80) were inconsistent, and the proportions of FGFR2 and MET amp were 1.
25% and 5%, respectively.
The corresponding proportion of metastases was 5% and 11%.

16 HER2-positive patients, 12.
5% (2/16) turned HER2-negative after treatment; Among the 64 HER2-negative patients, 3.
1% (2/64) became HER2-positive, indicating that the HER2 status of tumor tissues
in gastric cancer patients was also dynamically changed during treatment.

Figure 4.
Genomic heterogeneity in the PANGEA clinical trial of molecularly driven therapy

For patients with gastric cancer metastasis, ctDNA can effectively reflect the genomic characteristics of metastases, and the primary lesions and metastases of 28 patients with metastatic gastroesophageal gland were detected by cell-free DNA (cfDNA), and 36% (10/28) of patients had significant differences in the genomes of primary and metastases, and the treatment regimen of 9 patients was adjusted
accordingly.

In terms of target gene amplification, 85% (17/20) of patients had metastases and cfDNA results; Among patients with inconsistent genomic changes between primary and metastatic lesions, 87.
5% (7/8) of patients had consistent results of metastases and cfDNA, suggesting that metastatic or cfDNA sequencing is more effective than primary sequencing guidance
.