To the Editor.
—There are two questions that come to mind upon reading "Survival in Patients With Lung Cancer" by Senior and Adamson in the June issue (Arch Intern Med125:975-980, 1970).
I seriously question the validity of defining "time zero" as the day on which the tissue diagnosis of cancer was established. Although it is obviously a convenient reference point, I would suggest that by doing so one is minimizing one of the main points made in the study somewhat, namely asymptomatic neoplastic disease of the lung offers a more favorable prognosis than symptomatic disease. Time zero may indeed represent time X plux an indefinite length of time Y, time X being the asymptomatic and unrecognized time of onset of neoplastic lung disease in the patient. I am not aware of any retrospective studies done with respect to the presence of neoplastic lung disease prior to the