New Diagnostic Test for the Early Detection of Lung Cancer

I was invited to discuss a new diagnostic test for the early detection of lung cancer by Gerri Willis of Fox Business News’ Willis Report.
40-110-000-web
An Italian clinical study presented at the September 2014 European Respiratory Society described 82 patients with abnormal chest x-rays. Patients breathed into a machine that measured the temperature of the exhaled air. Forty of the patients ultimately proved to have cancer and 42 did not, as confirmed by subsequent biopsy. They found a correlation between the temperature of the exhaled breath and presence of lung cancer. They also found that long term smokers had higher breath temperatures, as did those with higher stage disease.

For a variety of reasons, a test as simple as breath temperature seems unlikely to be highly specific. After all, the temperature of the exhaled breath could reflect infection, inflammation, or even activity level, as vigorous exercise can raise the body’s core temperature. Nonetheless, the fact that there is any correlation at all is of interest.

PET scan lung cancerWhat might underlie these findings? Accepting the shortfalls of this small study, it is an interesting point of discussion. First, cancer is a hyper metabolic state. Cancers consume increased quantities of glucose, proteins, and lipids. PET scans measure these phenomena every day. Second, cancer is associated with hyper vascularity. Up-regulation of VEGF could cause hyperemia (increased capillary blood flow) in the airways of lung cancer patients, resulting in the finding. Finally, cancer, in and of itself, is an inflammatory state. Inflammation reflects increased metabolic activity that could manifest as a whole body change in basal temperature.

Lung cancer is the leading cause of cancer death in the US, constituting 27% of all cancer deaths. Despite the over 224,000 new diagnoses and 160,000 deaths, the five-year survival for lung cancer today at 17% has not changed in several decades. Nonetheless patients who are detected early (Stage I) have a greater than 50% five-year survival.

We know from the National Lung Cancer Screening Trial published in 2010, that early detection by CT scans can reduce mortality from this disease by 20%. In the cancer literature, that is huge. The problem is that screening CTs are comparatively expensive, inconvenient, expose patients to radiation and are themselves fraught with false positives and false negatives. Furthermore, it is estimated that that broad application of spiral CT’s could cost over $9 billion a year. Thus, simple, non-invasive screening techniques are sorely needed.

The use of exhaled breath to diagnose cancers has been under in development for decades. Recently, investigators from The Cleveland Clinic and others from Israel have reported good results with a microchip that measures the concentration of volatile organic compounds in the breath and provides a colorimetric score. With several hundred patients the receiver-operating curves (ROC, a technique that gauges the sensitivity and specificity of a test) in the range of 0.85 (1.0 is perfect) are quite favorable. Although these techniques have not yet gained broad application, they are extremely interesting from the standpoint of what it is they are actually measuring.

For decades, the principal focus of scientific exploration in cancer has been genomic. Investigators at Boston University and others at MD Anderson in Texas have used genomic and methylation status of oro-and naso-pharyngeal swabs to identify the earliest hallmarks of malignant transformation. To the contrary, the breath tests described above measure phenomena that fall more in the realm of metabolomics. After all, these are measures of cellular biochemical reactions and identify the transformed state at a metabolic level.

Though still in its infancy, metabolomics reflects the most appealing of all cancer analyses. Examining cancer for what it is, rather than how it came to be, uses biochemistry, enzymology and quantitative analyses. These profile the tumor at the level of cellular function. Like the platforms that I utilize (EVA-PCD), these metabolic analyses examine the tumor phenotype.

I applaud these Italian investigators for using a functional approach to cancer biology. This is a highly productive direction and fertile ground for future research. Will breath temperature measurement prove sensitive and specific enough to diagnose cancer at early stage? It is much too early to say, but at least for now, I wouldn’t hold my breath.

Circulating Tumor Cells and Early Diagnosis

A recent report describing a novel application of the cell search technology developed by Veridex, LLC (a subsidiary of Johnson & Johnson) may provide an extremely sensitive tool for the early detection of cancer. Four major cancer centers in the United States will conduct an analysis to determine the accuracy of this method for early diagnosis.

Over recent years, it has been recognized that cancer patients circulate small numbers of tumor cells in their blood. Using microbead technology, these tumor cells can be isolated from the blood stream and characterized. The original application of the technology was a prognostic marker by which patients with breast, colorectal or prostate cancers and high levels of circulating tumor cells, fell in the “high-risk” groups. I have been highly supportive of the application of this technology and have applied it extensively for patients with prostate and breast cancers.

The more recent iteration of this technique will allow investigators to not only identify but also characterize the isolated tumor cells. This provides an exciting new opportunity for early diagnosis.

As we speculate on the ramifications of this discovery, certain questions are raised. The most immediate being: What to do with the data? It has previously been suggested that many cancers arise 20 or 30 years before they are clinically detected. Malignant populations measuring in the hundreds of thousands, millions or even hundreds of millions, may still lie below the radar screen of modern diagnostic tools. If we have the capacity to identify patients 10 or 20 years before their cancers can be clinically detected, would we then begin therapy decades before clinical disease arises? If so, what treatments will we administer? Will the early detection of cancer cells be associated with the further characterization of tumors, such that targeted agents can be utilized to eliminate these clones at their earliest inception?

We will watch the development of these clinical studies with great interest. It will be even more interesting to see how we answer the questions that arise.