Take a sample from the suspected area, polyp, ulcer etc. Extract RNA from these cells. Convert to cDNA and do real time PCR to determine whether the oncogene VAV3 is over expressed? (VAV3 is highly linked to CRC).
Yes. There are several things one would need to consider very carefully before claiming VAV3 as a good diagnostic marker in CRC or designing the diagnostic test for CRC using the method you proposed. Here are a few that come to mind.
1. There seems to be only one paper that links VAV3 over-expression to CRC (http://www.nature.com/articles/srep09360).
2. This paper linked VAV3 immunohistochemical positivity in CRC vs normal colon using FFPE tissue, finding that VAV3 protein over-expression is an independent prognostic marker in CRC. I do not believe that they showed that the level RNA expression of VAV3 in CRC is correlated to over-expression of VAV3 protein. There are other means by which protein levels can be regulated. That should be tested first before one extrapolates correlation to RNA.
3. Don't think they looked at VAV3 expression in polyps. Is VAV3 over-expression in polyps correlated to development of CRC later on? At what level of over-expression (RNA and protein) is there an increased risk to develop CRC?
4. 11% of tumour samples in one of their patient cohorts were completely negative for VAV3 protein expression.
5. They split patients into two groups, low and high VAV3 protein expression levels and did Kaplan-Meier survival plots. High VAV3 patients did worse than low VAV3 patients. Can one detect a statistically significant difference in VAV3 RNA expression in CRC to stratify patients into these two groups? Will VAV3 RNA levels in polyps be significantly different enough from normal tissue to make a claim of the likelihood of CRC or possibility of CRC down the line. What are the thresholds?
6. Sample preparation/storage....fresh biopsy vs FFPE.
DNAses and RNases are very low or inexistant in tumours. See Taper HS et al. in J Histochem Cytochem and Cancer journals; old techniques with light microscopy but it worked well.
I would advise complete pathological review of the sample prior to taking areas of tumour. H+E sections plus a marker such as ki67 IHC would be a basic way of identifying truly cancerous areas. Also with heterogeneity it may make sense to collect multiple samples as VAV3 may not be universally expressed?
- Badges
- Science topic
- Similar topics
- Medicine
- Oncology
- Cancer Research