Broadly speaking, the goal of cancer screening is to shift disease detection to the furthest point upstream in the course of the neoplastic process at which treatment can prevent a precancerous lesion from becoming invasive, thus averting cancer, or prevent an early cancerous lesion from spreading, thus averting death from cancer. The challenges go beyond defining end points and choosing tests and procedures: not all precancerous lesions progress to invasive disease, and not all early invasive cancers progress to causedeath. The more upstream the lesion end point targeted for intervention, the greater the probability of overtreatment and harm to patients. Conversely, acting conservatively by shifting the trigger for treatment to downstream end points risks missing lesions that should have been treated earlier.

Anyone concerned with the challenges of screening for cancer, particularly from anatomical sites that are relatively inaccessible, must view the linked articles (doi:10.1136/bmj.b2546; doi:10.1136/bmj.b2548; doi:10.1136/bmj.b2549) from the trial of management of borderline and other low grade abnormal smears (TOMBOLA) as an embarrassment of riches.1 …