A study examines the role of different tumor markers or their combinations in diagnosing pancreatic cancer in patients with a solid lesion in the pancreatic head. The following table is reported: Based on the study results, a patient with a solid lesion in the pancreatic head is most likely to be free of pancreatic cancer if which of the following tests or combinations is negative?

Question

A study examines the role of different tumor markers or their combinations in diagnosing pancreatic cancer in patients with a solid lesion in the pancreatic head.  The following table is reported: Based on the study results, a patient with a solid lesion in the pancreatic head is most likely to be free of pancreatic cancer if which of the following tests or combinations is negative?

Quizplus · Accepted Answer

Explanation: The negative predictive value of a test is the probability of not having a disease given a negative result on that test. In this case, it is the probability of a patient with a solid pancreatic head lesion not having pancreatic cancer given a negative marker or combination of markers (parallel combination means all markers are drawn simultaneously, whereas serial combination means they are drawn in a specific order). Calculating the predictive value of a test requires knowing the disease prevalence, which is not given in this question. However, other measures can help evaluate the utility of the diagnostic tests presented. The likelihood ratio (LR) is an expression of sensitivity and specificity that can be used to assess the value of a diagnostic test indepedently of prevalence. The positive likelihood ratio (LR+) represents the value of a positive test result. It is the probability of a patient with the disease testing positive divided by the probability of a patient without the disease testing positive. The negative LR (LR-) represents the value of a negative test result. It is the probability of a patient with the disease testing negative divided by the probability of a patient without the disease testing negative. LR values can vary between 0 and infinity. In general, the smaller the LR, the less likely the disease presence (eg, LR = 5 increases probability of disease by ~30% whereas LR = 2 increases probability of disease by only ~15%); a LR < 1 argues against the disease (eg, LR = 0.1 decreases probability of disease by ~45%). LR In this case, the parallel combination has the lowest LR- (LR- = 0.03); therefore, negative test results on the parallel combination gives the strongest evidence against the presence of pancreatic cancer. The LR can be used to calculate post-test probability (post-test odds = pre-test odds * LR). The parallel combination test also has the highest sensitivity at 98.41%, meaning that patients with pancreatic cancer would have >98% probability of having a positive result on that test. Of all the tests and combinations listed, the parallel combination is the most likely to detect cases of pancreatic cancer and the least likely to give false negative results. Therefore, a negative result on this test is most helpful in ruling out pancreatic cancer in a patient with pancreatic head lesion. The very elevated sensitivity came at the cost of specificity, which was the lowest among the values reported. Educational objective: The negative likelihood ratio (LR-) represents the value of a negative test result. The smaller the LR, the less likely it is that the disease is actually present. A negative result on a highly sensitive test helps to rule out a disease. References: Diagnostic accuracy measures (http://www.ncbi.nlm.nih.gov/pubmed/24135733)

A Study Examines the Role of Different Tumor Markers or Their