Massive occurrences of interstitial loss of heterozygosity (LOH) likely resulting from gene conversions were found by us in different cancers as a type of single-nucleotide variations (SNVs), comparable in abundance to the commonly investigated gain of heterozygosity (GOH) type of SNVs, raising the question of the relationships between these two opposing types of cancer mutations. In the present study, SNVs in 12 tetra sample and 17 trio sample sets from four cancer types along with copy number variations (CNVs) were analyzed by AluScan sequencing, comparing tumor with white blood cells as well as tissues vicinal to the tumor. Four published “nontumor”-tumor metastasis trios and 246 pan-cancer pairs analyzed by whole-genome sequencing (WGS) and 67 trios by whole-exome sequencing (WES) were also examined. Widespread GOHs enriched with CG-to-TG changes and associated with nearby CNVs and LOHs enriched with TG-to-CG changes were observed. Occurrences of GOH were 1.9-fold higher than LOH in “nontumor” tissues more than 2 cm away from the tumors, and a majority of these GOHs and LOHs were reversed in “paratumor” tissues within 2 cm of the tumors, forming forward-reverse mutation cycles where the revertant LOHs displayed strong lineage effects that pointed to a sequential instead of parallel development from “nontumor” to “paratumor” and onto tumor cells, which was also supported by the relative frequencies of 26 distinct classes of CNVs between these three types of cell populations. These findings suggest that developing cancer cells undergo sequential changes that enable the “nontumor” cells to acquire a wide range of forward mutations including ones that are essential for oncogenicity, followed by revertant mutations in the “paratumor” cells to avoid growth retardation by excessive mutation load. Such utilization of forward-reverse mutation cycles as an adaptive mechanism was also observed in cultured HeLa cells upon successive replatings. An understanding of forward-reverse mutation cycles in cancer development could provide a genomic basis for improved early diagnosis, staging, and treatment of cancers.

Background: Cancer subtyping has mainly relied on pathological and molecular means. Massively parallel sequencing-enabled subtyping requires genomic markers to be developed based on global features rather than individual mutations for effective implementation. Methods: In the present study, the whole genome sequences (WGS) of 110 liver cancers of Japanese patients published with different pathologies were analyzed with respect to their single nucleotide variations (SNVs) comprising both gain-of-heterozygosity (GOH) and loss-of-heterozygosity (LOH) mutations, the signatures of combined GOH and LOH mutations, along with recurrent copy number variations (CNVs). Results: The results, obtained based on the WGS sequences as well as the Exome subset within the WGSs that covered ~ 2.0% of the WGS and the AluScan-subset within the WGSs that were amplifiable by Alu element-consensus primers and covered ~ 2.1% of the WGS, indicated that the WGS samples could be employed with the mutational parameters of SNV load, LOH%, the Signature α%, and survival-associated recurrent CNVs (srCNVs) as genomic markers for subtyping to stratify liver cancer patients prognostically into the long and short survival subgroups. The usage of the AluScan-subset data, which could be implemented with sub-micrograms of DNA samples and vastly reduced sequencing analysis task, outperformed the usage of WGS data when LOH% was employed as stratifying criterion. Conclusions: Thus genomic subtyping performed with novel genomic markers identified in this study was effective in predicting patient-survival duration, with cohorts of hepatocellular carcinomas alone and those including intrahepatic cholangiocarcinomas. Such relatively heterogeneity-insensitive genomic subtyping merits further studies with a broader spectrum of cancers.

To delineate the temporal sequence of mutational events, residue-by-residue comparisons of same-patient tetra samples comprising white-blood-cell control, tumor-adjacent morphologically largely normal ‘subnormal’ and ‘paratumor’ regions, and tumor tissue were analyzed. Cells in the ‘subnormal’ region were enriched in single-nucleotide-variations (SNVs), including both gain-of-heterozygosities (GOHs) and loss-of-heterozygosities (LOHs), and copy-number-variations (CNVs); the CpG>TpG GOHs and TpG>CpG LOHs represented the most prominent SNVs especially in early-onset patients, and were found to be located close to recurrent CNVs. A striking fraction of the SNVs and CNVs displayed by the ‘subnormal’ region cells were reverted in the immediately tumor-proximal ‘paratumor’ region cells back to their pre-mutated control genotypes with strong lineage effects particularly in the LOH-reversals. These findings, corroborated by results from previously-reported tumor genomic changes analyzed using whole-exon-sequencing and whole-genome-sequencing, and from sequential samplings of cultured HeLa cells, led to the proposal of a ‘Stage Specific Population’ (SSP) model of carcinogenesis, where cancer development proceeded from control cells to ‘subnormal’-stage cells followed by ‘paratumor’-stage cells and finally the mature tumor cells. On this basis, the ‘subnormal’ and ‘paratumor’ stages, corresponding respectively to a premalignant stage and a premalignant or early malignant stage, are identified as plausibly valuable targets for therapeutic intervention.