Oncology: ColonoKit

Colorectal Cancer (CRC) incidence and the impact of detecting it early, make CRC diagnosis a primary focus in the oncology community. ColonoKit is a screening solution for colorectal cancer, a cancer which early diagnosis is crucial for patient survival. It will constitute a revolution in colorectal cancer diagnosis since it jointly displays 3 very attractive features:

  • A very high sensitivity and a very high specificity
  • High acceptance by the patient (based on a blood sample)
  • Limited cost

Below, you can learn more about colorectal cancer, its diagnosis, and ColonoKit.

Disclaimer: ColonoKit is an innovation by DNAlytics which is not yet granted with market access.

Colorectal Cancer

Colorectal cancer (CRC) is cancer of the lower part of the digestive system (large intestine). It generally starts with the growth of polyps. Polyps are non-cancerous (benign) heaps of cells on the internal part of the colon. Within a few years, some polyps might get bigger and turn into malignant (cancerous) tumors and spread first locally than globally. The causes of CRC, as many other cancers, are unknown, excepted in a few cases where the cause might be the inheritance of a genetic mutation. Statistical risk factors have however been identified. One of them is of being over 50 years of age. In the United States for example, the NIH’s National Cancer Institute Surveillance, Epidemiology and End Results (SEER) program1 provides the following statistics for 2015:

  • 42.4 new CRC cases appear each year per 100.000 individuals (men and women). This figure varies between ethnicities and gender, but is in any case not lower than 30/100.000 (hispanic women) and goes up to 61.2/100.000 (black men).
  • 4.5% of men and women will be diagnosed with CRC at some point during their lifetime
  • 82% of the cases of CRC occur between 45 and 84 year.
  • CRC is the second cause of cancer-related death (after lung and bronchus cancer)
  • 1 cancer case over 12 is a CRC

This shows how important CRC is. Furthermore, early CRC is symptomless, and thus frequently diagnosed when already advanced. But metastatic disease is associated with a very poor prognosis. In contrast most patients can be cured by early tumor resection : Only 13.1% of patients diagnosed too late (i.e. once metastases are already present) survive beyond 5 years. But on the contrary 90.1% of the early diagnosed patients survive beyond 5 years. Early diagnosis is thus a crucial factor for favorable disease management.

Colorectal Cancer diagnosis

The most widespread techniques for CRC diagnosis are colonoscopy and stool tests. As age is a recognized risk factor, and given both the incidence and the poor prognosis of this cancer, several countries have included in their national health policy a recommendation for a systematic CRC screening, for example every other year between 50 and 70 years old for everyone. But the existing solutions fail at generating a high level of compliance among the patients, because of the costs and/or of the inconvenience of the tests.

Colonoscopy is an invasive examination. It is the gold standard for CRC detection, but not used as a systematic screening test because of its high cost and the low available resources in hospitals.

  • Specificity: 100%
  • Sensitivity: 98%
  • Cost: between a few 100’s and 2500€
  • Compliance of the patients: low

FIT – Faecal Immunochemical Test is an immunochemical test on stools. It aims at detecting traces of blood in stools. It requires several stool samples and has to be repeated for one application of the test. Given its low cost, it is used as a screening test in some countries. However, it misses a lot of patients with CRC (low sensitivity) and also raises a number of false alarms (limited specificity).

  • Specificity: 89%
  • Sensitivity: 66%
  • Cost: ~60€
  • Compliance of the patients: Low (40% estimation)

FIT is itself an improved version of the FOBT –Faecal Occult Blood Test, which is an immunochemical test on stools as well, requiring up to 6 stool samples and several repeats for completing the test just once. Its use might on top of that also require dietary preparation. Given its very low cost (even lower than FIT) it is used as a screening test in some countries, although there are doubts as if the test really captures early stages.

  • Specificity: 88%
  • Sensitivity: 50-60%
  • Cost: ~20€
  • Compliance of the patients: Low (40% estimation)

None of these solutions are well accepted by the target population. The socio- economic burden of these procedures is also very high. Other approaches exist but are not validated or well accepted or only provide poor results. Thus, there is urgent need for a sensitive and specific, non-invasive solution for early CRC diagnosis to avoid disease progression to advanced stages that are difficult to cure. ColonoKit, described in the next section, addresses that need.

About ColonoKit

ColonoKit is based on the measurement of a series of 22 transcriptomic markers (mRNA) in a specific subpopulation of cells (monocytes) extracted from simple blood samples. The extraction of these particular cells can be performed manually and fast in clinical biology laboratory within most hospitals based on simple devices capturing monocytes specifically. The procedure has been standardized and the consumables required can be purchased from very well-known providers. The sole constraint is to extract these cells within a limited time frame after blood sampling.

Using a transportation kit designed and validated by DNAlytics, the samples are brought to a central laboratory for mRNA extraction, and a genomic experiment is performed to assess the level of expression (“activity”) of the gene signature components.

These laboratory results are fed into a mathematical model that predicts the status of the patient: cancerous or not. If the prediction is positive (cancer) the patient is redirected to colonoscopy, which is still the gold standard nowadays.

Benefits

The advantages of ColonoKit are the following:

  • Performance: so far, the average between sensitivity and specificity is of 90%, while the blood tests (FIT and FOBT) show a value of between 50 and 80% for the same statistics. In other words, ColonoKit outperforms these solutions in terms of the number of cancer cases it discovers but it is also much more reassuring as it does not raise as many false alarms as these previous tests.
  • Acceptance: having blood sampled is much more accepted than a colonoscopy if it can be avoided, and it is also much more convenient than spreading stools multiple times on a scoring sheet. As a result, the compliance of the patient to the test should be high, and in particular higher than the compliance for recurring colonoscopy and stool tests.
  • Cost: ColonoKit is likely to be more expensive than stool tests, but will still represent only a small fraction of the cost of a colonoscopy.

These three features together make ColonoKit a solution that will be both preferred by the patient and attractive for healthcare systems.

Clinical Evidence

Overall, about 550 samples have been prospectively collected for this project so far.

In a first step, the gene signature has been identified on about 100 individuals (healthy volunteers and CRC patients) based on high-throughput transcriptomic data.

In a second step, the performance of the predictive solution has been assessed on about 200 samples, among which half from healthy volunteers and half from CRC patients. On these 200 samples, the predictive solution shows an AUC of 92%+/-4% and a Balanced Classification Rate (average between sensitivity and specificity) of 89%+/-5%.

In a third step, about 150 samples have been used for the validation of the same approach in a technological setting fitting routine requirements.

The remaining samples represent controls from other cancer types, treated patients or technical variations.

Extra clinical validation is still ongoing.