Side effects are carefully assessed and recorded. Patients are also observed closely for anticancer effects by
  repeated measurement of the cancer sites present at the beginning of the trial. If those sites shrink appreciably,
  the patient is said to have responded to the treatment. Phase I treatments may produce anticancer effects and
  some patients have been helped by them.
• Phase II  These studies determine the effect of a research treatment on various types of cancer. Usually,
  groups of 30 to 40 patients with one type of cancer—breast cancer that has become resistant to standard
  therapy, for example—receive a Phase II treatment. Side effects are assessed and, since more people receive
  the treatment than in Phase I, there is more chance to observe less common side effects. Patients are closely
  observed for anticancer effects, and response rates are determined. If at least one-fifth of the patients respond,
  the treatment is judged to be active against the tumor type. A treatment that shows activity against cancer in
  this phase moves to Phase III.
• Phase III  Trials in this phase require large numbers of patients—some trials enroll thousands—who are
  divided into groups. One group, called the "control" group, may receive the standard accepted treatment so the
  new treatment can be directly compared to it. The control group may receive the chemotherapy usually given
  for a certain type of cancer, for example, while another group receives the new treatment to see if it improves
  survival. All patients in Phase III trials are monitored closely for side effects, and treatment is discontinued if
  the side effects are too severe.