For the best experience on the Abcam website please upgrade to a modern browser such as Google Chrome
If you continue without changing your cookie settings, we'll assume you’re happy with this.
At the core of tackling cancer is diagnosing patients as soon as possible, as survival rates can drop as the disease progresses. For patients with non-small-cell lung cancer (NSCLC) – the most common form of lung cancer – for example, the five-year survival rate for Stage I cancer can be over 90%.1 But this rate of survival can drop to as low as 1% in those patients with stage IV NSCLC. Since a cancer diagnosis usually only happens when it becomes symptomatic, this increases the chance that it may already be metastatic and invasive, and so, potentially untreatable. A rapid and accurate diagnosis at the earliest possible stage of cancer is therefore a top priority for many cancer researchers.
A team of researchers from the University of Maryland has taken steps towards achieving a method of early diagnosis for lung cancer. Looking at cell-free microRNAs (miRNAs) circulating in the blood from a small group of NSCLC patients, the team set out to use multiplex techniques to develop plasma biomarkers for the detection and histological classification of this lung cancer.2 This approach would minimize the need for invasive tissue samples and offer an early warning system for NSCLC.
MicroRNAs and the liquid biopsy
Non-coding RNA (ncRNA), including miRNA, makes up 60–70% of the total human genome3. and miRNAs have a broad range of biological function through their ability to regulate gene expression – usually by binding to the 3’ UTR of mRNA.
A rapid and accurate diagnosis at the earliest possible stage of cancer is a top priority for many cancer researchers.
The specific set of miRNAs circulating in the blood of a person with cancer can be subtlety different from those in a healthy patient. This is because tumor cells release distinct miRNAs into the bloodstream, which can offer vital information on a cancer’s subtype, progression, and even therapy resistance.4
This makes miRNAs from plasma a logical choice for liquid biopsy, which is convenient and less invasive than taking a tissue sample and could provide both diagnostic and prognostic power. A liquid biopsy also allows clinicians to take repeated samples, making it possible to monitor circulating miRNAs during cancer progression and treatment.
Multiplexing miRNAs with FirePlex
The researchers opted to use FirePlex to quantify circulating miRNA. FirePlex technology is a relatively new multiplex approach that relies on hydrogel particles and custom miRNA probes for detection. It’s commonplace to use quantitative reverse transcription polymerase chain reaction (qRT-PCR) to quantify miRNAs, although this approach can only look at a limited number of miRNAs simultaneously.
FirePlex was easier, quicker, and cheaper than conventional qRT-PCR with equivalent analytical performance.
Another option is to use microarrays, which are capable of analyzing more miRNAs but suffer from lower detection sensitivity, making detection of low-abundance miRNAs a problem. Microarrays and qRT-PCR together require large sample plasma volumes from which to purify RNA, followed by additional steps that include RT and sequencing. Conducting sequencing itself can be a long process and comes with its own challenges like variations in data due to library preparation methods.
FirePlex alleviates the majority of these issues. Requiring as little as 10 mL sample from biofluids, it bypasses the need for RNA purification and RT – this minimizes the chance of errors during sample preparation. The researchers from the University of Maryland study said that FirePlex was easier, quicker, and cheaper than conventional qRT-PCR with equivalent analytical performance.
From a total of 56 mixed-stage NSCLC patients and 28 cancer-free patients who smoked, the team of researchers used FirePlex to quantify 11 lung tumor-associated miRNAs from 20 mL of plasma. The team had already made use of next-generation deep sequencing to identify and characterize these 11 miRNAs from lung tumor tissue as potential plasma biomarkers for lung cancer.
To test the performance of the FirePlex method, qRT-PCR was run alongside on plasma samples from 20 lung cancer patients and 20 cancer-free controls to quantify miRs-34a-5p. qRT-PCR required 200 mL of plasma, RT to generate cDNA and PCR to measure target gene expression levels. The statistically significant results from this head-to-head found comparable expression levels and saw no difference in the coefficient of variation (CV) of miRNA results between FirePlex and qRT-PCR.
The analyses of these patient liquid biopsies show how FirePlex has excellent intra-assay consistency, high sensitivity, and improved sample-usage compared to qRT-PCR. FirePlex also costed less than qRT-PCR – due to the lack of RNA isolation and RT – and took 3–5 hours from sample to data, compared with the 24-hour workflow of qRT-PCR.
Early detection and classification of lung cancer
Of the 11 miRNAs looked at, the team found ten that had different expression levels in cancer vs. cancer-free patients, meaning these were a very promising set of miRNAs to use as plasma biomarkers for NSCLC.
The work here represents a non-invasive tool for both the early detection of lung cancer and the classification of its subtype
With a complex set of statistical analyses, they managed to hone in two miRNAs in particular: miR-205-5p and miR-210-3p. This pair of miRNAs seemed to have as much diagnostic power as the whole set of 10 initially identified when used in combination. Previous work has already shown that elevated expression of miR-205-5p can contribute to NSCLC progression,5 while hypoxia-induced miR-210-3p regulate the susceptibility of cancer cells to lysis by cytolytic T cells.6
From just miRs-205-5p and -210-3p, the team devised a prediction model with 78.6% sensitivity and 89.3% specificity for identifying lung cancer – a value independent of tumor stage, and the patients’ age and sex. With this same pair of miRNAs, they went from detecting lung cancer to classifying it as either squamous cell carcinoma or adenocarcinoma. These models successfully delivered 75.0% sensitivity and 89.3% specificity for a diagnosis of lung squamous cell carcinoma, and 82.2% sensitivity and 89.3% specificity for lung adenocarcinoma.
Given the similar sensitivity and specificity of the models in both early and late-stage cancer, the work here represents a non-invasive tool for both the early detection of lung cancer and the classification of its subtype – thereby helping clinicians select the most appropriate therapy.