Objective: To analyze poly-
guanine (
poly-G) genotypes and construct the phylogenetic tree of
colorectal cancer (CRC) and provide an efficient and convenient method for the study of intra-
tumor heterogeneity and
tumor metastasis pathway. Methods: The clinicopathological information of patients with primary
colorectal cancer resection with regional
lymph node metastases were retrospectively collected in the Department of General Surgery, General Hospital of Tianjin Medical University from January 2017 to December 2017. The
paraffin sections of the paired
tumor samples were performed consecutively, and multi-region microdissection was performed after histogene staining. The
phenol-
chloroform extraction and
ethanol precipitation scheme was used to obtain
DNA, and
Poly-G multiplex PCR amplification and capillary electrophoresis detection were performed. The correlation between
Poly-G mutation frequency and clinicopathological parameters was analyzed. Based on the difference of
Poly-G genotypes between paired samples, the distance matrix was calculated, and the phylogenetic tree was constructed to clarify the
tumor metastasis pathway. Results: A total of 237 paired samples were collected from 20 patients including 134 primary lesions, 66
lymph node metastases, 37 normal tissues, and
Poly-G mutation was detected in 20 patients (100%). The mutation frequency of
Poly-G in low and undifferentiated patients was (74.10±23.11)%, higher than that in high and medium differentiated patients [(31.36±12.04)%, P<0.001]. In
microsatellite instability patients, the mutation frequency of
Poly-G was (68.19±24.80)%, which was higher than that in microsatellite stable patients [(32.40±14.90)%, P=0.003]. The
Poly-G mutation frequency was not correlated with age, gender, and pathological staging (all P>0.05). Based on
Poly-G genotype difference of the paired samples, the phylogenetic trees of 20 patients were constructed, showing the evolution process of the
tumor, especially the subclonal origins of
lymph node metastasis. Conclusion:
Poly-G mutations accumulate in the occurrence and development of CRC, and can be used as
genetic markers to generate reliable maps of intratumor heterogeneity in large numbers of patients with minimal time and cost expenditure.