The study covered in this summary was published in medRxiv.org as a preprint and has not yet been peer reviewed.
Key Takeaways
Researchers have developed and validated a new screening test to detect pancreatic cancer in people with newly diagnosed diabetes.
The new test relies on detecting a very distinctive epigenetic marker of pancreatic cancer in the blood, the appearance of 5-hydroxymethylcytosine (5hmC) in cell-free DNA that’s isolated from a 20 mL blood specimen.
This cost-effective technology will be useful in detecting occult pancreatic cancer at an early stage in high-risk patients, which may greatly improve survival.
Why This Matters
Pancreatic cancer has the worst overall survival of all major cancers, with a 5-year relative survival rate of 10%.
The majority of patients with pancreatic cancer have few, if any, symptoms at the early stages; therefore, they are generally not diagnosed until late stages when there have already been metastases, at which point it is often too late for curative surgery.
New-onset diabetes is a risk factor for pancreatic cancer; more than 50% of patients with pancreatic cancer have a known history of diabetes mellitus.
The authors concluded that the screening tool they developed and validated is precise, scalable, and cost-effective, and that it is a promising way to allow early detection of occult pancreatic cancer in higher-risk people such as those with new-onset diabetes.
This new test has the potential to improve survival of people with new-onset diabetes because earlier diagnosis of pancreatic cancer facilitates an early start to cancer treatment.
Study Design
The derivation phase of the study used blood specimens from 89 people with pancreatic cancer and 596 people without cancer.
The validation phase included blood specimens from 79 people with pancreatic cancer, 163 people without any form of cancer, and 506 people with a non-pancreatic type of cancer (35 prostate, 97 breast, 146 colorectal, and 228 lung).
A second validation assessment involved specimens from 91 people with new-onset diabetes, including 29 with a diagnosis of pancreatic cancer and 62 without cancer.
Study subjects were age 45-75 and enrolled at any of 146 US sites.
The test detected 5-hmC in cell-free DNA in plasma specimens isolated from 20 mL of a person’s blood
The investigators used machine learning to develop a predictive model for detecting pancreatic cancer based on the 5hmC signatures found in the cell-free DNA from each subject.
Key Results
In the derivation phase the test model showed a mean sensitivity of 61.1% and mean specificity of 97.6%.
The derived prediction model could account for (had an area under the receiver operating characteristic curve of) 87%-96% of pancreatic cancers, with the specific value varying depending on the cancer stage, I-IV.
When tested on patients with pancreatic cancer (42% with early-stage cancer) and controls without cancer the screening test showed sensitivity of 51.9% and specificity of 100%.
Similar results occurred in patients with new-onset diabetes, in whom the test had 55.2% sensitivity and 98.4% specificity.
The 5hmC signatures that signaled pancreatic cancer were similar between patients with new-onset diabetes and those without diabetes, suggesting that the test can detect pancreatic cancer in patients with or without diabetes.
Implied Limitations
The authors did not identify any study limitations in their preprint.
Study Disclosures
This study was funded by Bluestar Genomics. Bluestar Genomics was involved in the study’s design, study execution, analysis, and interpretation.
All authors are full-time employees and shareholders of Bluestar Genomics.
This is a summary of a preprint research study, “Validation of a Pancreatic Cancer Detection Test in New-Onset Diabetes Using Cell-Free DNA 5-Hydroxymethylation Signatures,” written by authors from Bluestar Genomics on medRxiv and provided to you by Medscape. This study has not yet been peer reviewed. The full text of the study can be found on medRxiv.org.
For more diabetes and endocrinology news, follow us on Twitter and Facebook.
Source: Read Full Article