Sporadic Colorectal Tubular Adenomas Thrive in Symbiosis With Underlying Nondysplastic Branching Crypts

Discussion

This study in German patients showed that NDCB -mostly NDCAB- frequently developed beneath the adenomatous tissue of TA. Similar observations were previously found in Swedish patients with TA (12, 13). The similarity between the two populations, who reside in different geographic areas with different microenvironments, suggests that NDCB may be a ubiquitous, unreported phenomenon in TA.

The frequency of NDCB correlated well with increasing adenoma size, suggesting that the increased size of TA was a consequence of the increased frequencies of both NDCB and DCB (14) in individual TA. In contrast, the frequency of NDCAB/NDCB was unaffected by clinical attributes, such as age, sex, or TA localization, indicating that the production of NDCB was not choreographed by hormones like estrogen or testosterone, by increasing age, or by TA localization in the colon or rectum.

Considering such findings, several key questions seem relevant:

1. What is the significance of NDCB beneath TA?

The normal colorectal mucosa in adults has a calculated surface of 2 m2 (16). Along that vast mucosal area only occasional crypts in symmetric branching are detected and crypts in asymmetric branching do not occur (12). It is therefore remarkable that haphazardly distributed TA hubs measuring only few mm, coexist with NDCAB underneath. Together, these histological findings suggest that in TA, both NDCB underneath and the dysplastic crypts on top (14) may be interdepended, with the theoretical possibility of a biological crosstalk between the two compartments.

2. Are colorectal NDCB hubs centers of field cancerization? Previous investigations on nonpolypoid colon adenomas (17) and polypoid colorectal adenomas (12) in Swedish patients showed that mucosal hubs displaying NDCSB and NDCAB were often present underneath the neoplastic canopy of conventional adenomas. Based on those findings and on the findings in point 1 (vide supra), it was previously postulated that nondysplastic mucosal hubs, particularly those exhibiting NDCAB could be putative cores of field cancerization (12). The present findings in German patients support that assertion. The finding of mucosal hubs exhibiting NDCB underneath colon adenomas elicited in laboratory animals by the aid of colonotropic carcinogens (10, 11) substantiates the findings in humans.

3. Are NDCB necessary structures from which the dysplastic epithelium in TA eventually develops?

In a recent study of the adenomatous tissue in 139 TA in German patients (14) a total of 3,956 DCB were found: 98% with asymmetric branching (DCAB) and the remaining 2% with symmetric branching (DCSB) (14). Those findings together with the present ones, reveal that nondysplastic and dysplastic crypts with asymmetric branching are the most common phenotype of branching crypts in both TA compartments.

4. Is the increasing frequency of NDCB in TA an indication of crypt proliferation?

The word “proliferation” derives from French” prolifère” (of flowers whose pedicel continues to grow above the pistil) and from Latin” proles”, meaning producing offspring.

The notion of crypt proliferation (i.e., branching crypts yielding eventually two or more offspring), is at variance with that of “cell proliferation”, often used to denote DNA synthetizing cells, knowing that offspring-crypts may or may not crystallize. We have applied the term cell proliferation (18) when cells in mitoses can be demonstrated (19).

In the present work, the notion of “crypt proliferation” is applied when crypt branchings were present.

5. Should the epithelium lining NDCB (that is NDCSB and NDCAB), be referred to as “normal” or as “nondysplastic”? Based on the knowledge gained on cell mutations in normal cells (vide infra, point 6), we opted to define the apparent histologically normal epithelium lining NDCB “nondysplastic”.

6. Are NDCAB the histologic smoking gun of somatic mutations?

In 1990 Fearon and Vogelstein (20) translated histological images of the adenoma-carcinoma sequence (5) into successive somatic genetic mutations resulting in the clonal expansions of critical genes.

In 2003, Yang et al. found that chromosome segregation defects contributed to aneuploidy in normal neural progenitor cells (21). Subsequent analyses of embryonic neural progenitor cells revealed that about 33% of the normal neural progenitor cells were aneuploid (22). In 2014, Jaiswal et al. (23) analyzed whole-exome sequencing data from DNA in the peripheral-blood cells of 17,182 persons who were unselected for hematologic phenotypes. They detected up to 18% clonal mutations in normal hematopoietic cells. The presence of a somatic mutation was associated with an increase in the risk of hematologic cancer (23). Sequencing studies of normal blood and skin revealed burdens and signatures of somatic mutations broadly like those observed in cancers from those cell types. In blood, driver mutations with typical leukemia patterns were found in ~10% of individuals older than 65 years of age without leukemia. Individuals carrying these driver mutations have an elevated future risk of blood cancers (24).

In 2015, Tomasetti C, Vogelstein B, Parmigiani G. reported that more than half of the mutations present in some cancers were also present in the normal cells, where cancer had originated (25). Their conclusions were that those somatic mutations would have been present even if the tumor had not formed (25). In 2018, Kondrashov claimed that normal cells of different tissues can accumulate hundreds to a few thousands of substitutions without acquiring hypermutation status (26). The approximate range of human germline mutation-rates is 1.0-1.2×10⁻⁸ per nucleotide per generation. Although most mutations have no harmful effects, up to 10% of these mutations may be deleterious (26).

In summary, in the normal intestinal epithelium, the somatic mutation rate is in the order of 2 to 10 mutations/diploid genome/cell division. That mutation rate is not far from that in cancer cells. Aging is accompanied by mutation of normal cells (27). Lee et al. found, in ≥2,000 colonic crypts, that approximately 1% of normal colorectal crypts in middle-aged individuals had a driver mutation (28). Spontaneous mutations in somatic cells were found to accumulate during lifetime (29). Evidently, normal colonic cells, including aberrant NDCAB, are amenable to carry somatic mutations. From the above studies in normal cells, it becomes apparent that NDCAB may be the histologic smoking gun of somatic mutations in normal colonic cells that we have been searching for. The present histologic findings in TA in German patients are consistent with earlier histologic findings in TA in Swedish patients (12) and with those in experimental animals (10, 11) thereby supporting the idea that the NDCAB are mutated aberrant structures upon which TA may eventually develop.