Using mass spectrometry to diagnose cancers and non-cancerous diseases

Using mass spectrometry, Emory University scientists have identified specific protein fingerprints in the cerebrospinal fluid (CSF) of humans that enable them to differentiate between cancers, both primary and metastatic, and non-cancerous diseases of the central nervous system (CNS). This finding offers promise of a more rapid, less invasive, and more accurate method of diagnosis of CNS diseases, especially the difficult-to-diagnose CNS cancers, and a better way to monitor disease progression and/or response to therapy.

"Because CSF sampling through lumbar puncture is less invasive than neurosurgical procedures for obtaining tissue, CSF is a desirable early step for establishing a diagnosis," says Dr. Savvas E. Mendrinos. "The diagnostic capabilities afforded by new proteomics technology give CSF sampling the potential to be an extremely useful tool for differentiating among the different causes of CNS disease and for monitoring disease progression and response to therapy."

Mass spectrometry is one of the primary tools of proteomics, a rapidly developing field in which scientists identify and characterize proteins and protein fragments. The study is presented by Dr. Mendrinos at Experimental Biology 2004, as part of the scientific sessions of the American Society of Investigative Pathology. Other authors, also from Emory University School of Medicine, are Dr. Melinda M. Lewis and Dr. Daniel J. Brat.

Because the brain and spinal cord are bathed in CSF, diseases of the central nervous system often are accompanied by cellular, protein, and chemical changes in the CSF. Different CNS diseases would be expected to result in characteristic alterations in the cellular and protein content of the CSF. Traditional cytopathologic examination of the cells within the CSF under the microscope, however, often results in nonspecific findings, especially in the diagnosis of primary and metastatic cancers. Even in the presence of positive clinical and radiographic findings, cytopathologic findings are positive in less than half of patients, and most of these positive results do not provide a definitive diagnosis. Therapy is sometimes delayed while physicians await the results of repeated lumbar punctures, which can continue to yield negative results even in the face of positive clinical findings. Neurosurgical procedures are more definitive but much more invasive than lumbar punctures.

The Emory research team used SELDI-TOF mass spectrometry to analyze the composition of proteins in the CSF of 30 patients, including 10 with known malignancies (primary glioblastomas and metastatic carcinomas); 12 with non-cancerous conditions including infections, multiple sclerosis and stroke; and 8 patients with no specific neurologic disease, only mild cognitive impairment. They identified specific protein fingerprints in CSF that reliably distinguished between cancerous and non-cancerous disease. A mass spectrometry profile containing three specific protein variations was always present in the CSF of cancer patients, but was never present in the CSF of patients without cancer. A completely different protein profile including four protein variations was present in the CSF of patients with non-cancerous diseases.

The next step in the ongoing research will be to use mass spectroscopy to identify and characterize the specific proteins that relate to each different disease category. The investigators then plan to test the diagnostic sensitivity of the protein profiles for larger scale clinical use.