As principal investigator Dr. Stephen Lam and colleagues explain in the Journal of Clinical Oncology published online on May 4, their strategy “used the same subject as his/her own control to identify elevated proteins in the pulmonary venous effluent draining the tumor vascular bed compared to matched systemic arterial blood.”
“The analytic issue,” they add, “is reduced to determining what is changed in an individual pre- and post-passage through the affected organ to get around the problem of finding low abundance markers in blood.”
In the discovery phase of their study, Dr. Lam, from the British Columbia Cancer Agency in Vancouver, and his team obtained paired samples of pulmonary venous blood and radial arterial blood from 16 patients with lung cancer.
Mass spectrometry analyses showed that levels of connective tissue-activating peptide III (CTAP III) and haptoglobin (a previously identified biomarker) were significantly increased in venous blood. Their findings were confirmed in a cohort of 64 lung cancer patients.
When they measured levels of the two proteins in plasma samples from 28 patients with lung cancer prior to and following surgical resection, levels of CTAP III decreased by a mean of 57% after surgery (p = 0.01). However, five patients experience a recurrence of lung cancer, and re-analysis showed that only the subset of patients who remained disease-free had significant reductions in CTAP III after the initial surgery (p = 0.002).
In two separate, population-based cohorts, levels of CTAP III were elevated among those later found to have lung cancer, predating diagnosis by up to 29 months.
In a multimodal model that included CTAP III, haptoglobin, age, smoking status, and FEV1% predicted, the positive predictive value of CTAP III was 62.7% and the negative predictive value was 88.5%.
“Our study thus underscores the importance of applying blood biomarkers not as a stand-alone test but as part of a multimodal risk lung cancer prediction model,” the authors conclude.
In a related editorial, Dr. Mark W. Duncan of University of Colorado Denver, calls this approach “an intriguing idea.” He adds, however, “The marked heterogeneity of lung cancer makes it highly unlikely that the full spectrum of this disease can be reflected in a couple of proteins.”
J Clin Oncol 2009;27.