IN ALL OF clinical medicine, few situations pose

as many heart-rending choices for patient, family and doctor as the terminal stages of cancer. It is the time when everything known to be effective against this still utterly mysterious disease has failed, when a doctor must either do nothing or venture into the high-risk world of experimental treatments, and when a patient often must choose between loss of hope or another round of expectation followed by crushing disappointment.

Aspects of this world were explored in The Post's series, "The War On Cancer," last week. The series took on a valid and difficult subject and performed a service by launching a broad public discussion on an issue of great concern. It dealt with early clinical testing--so-called Phase I trials--of potential anti- cancer drugs. The series criticized these tests, which are run by the National Cancer Institute, for being designed primarily to determine drug toxicity rather than efficacy, for inflicting excessively toxic side effects on patients, including 620 cases of drug- related death, for errors in calculating proper dosage and sloppy record-keeping, and for continued testing of drugs "even years after studies failed to show that they were of use." It described drugs "derived from a list of highly toxic industrial chemicals including pesticides, herbicides and dyes" that have created a "litany of death and suffering" in "human experiments" where patients become "laboratory subjects." Most seriously, the series questioned whether the tests violate the physician's guiding tenet "first do no harm."

The pain endured by patients who agree to participate in experimental drug testing is often great, but must always be understood in the light of three fundamental facts. First, such patients are terminal; their life expectancies are generally a few weeks or months, and every known therapy has been exhausted. Second, because the causes of cancer are still unknown, doctors and drug designers must rely on the simple observation that cancer cells grow and divide where normal cells would not. This means that agents that prevent cells from growing or dividing will kill cancer cells faster than they kill normal ones. Thus chemotherapy, like other cancer treatments is, by definition, a toxic process. Finally, cancer, despite its name, is not one disease, but dozens of different ones.

Many of the otherwise inexplicable aspects of Phase I testing described in The Post's series are understandable in this light. Drugs must be tested on a variety of tumors before being discarded as ineffective. Therefore a drug may still be being tested years after early tests have shown no benefit. As with any type of new drug, doctors must find the right dose level and schedule of administration by a process of informed trial and error; there is no other way. In these early stages of testing, even the best drugs will be more toxic and less effective than later on when used in the optimal dose, in combination with other drugs, and against earlier, more susceptible stages of the disease.

Though mistakes have surely been made in record-keeping and reporting of results, the record for experimental chemotherapy does not appear to be any better or worse than in other areas of medicine. Most important, an experimental drug's low chance of success and high risk of causing toxic side effects must be balanced against the patient's high risk of death. As some of today's letters indicate, many patients are also performing the hardest balancing act of all: weighing their own low chance of benefit against their contribution to those who come after.

There are serious issues relating to experimental chemotherapy. The NCI's large clinical testing programs may be shortchanging research programs, in pathology, toxicology and pharmacology, that could make cancer treatment more effective and less painful. Some doctors may be losing sight of the individual patient's welfare in their zeal for research. But finding the middle ground between the quality of life for a dying patient and the possibility of finding a cure for others is not easy. For some patients, the best thing is do something, even something painful. For others it will be different. Ultimately these choices depend on the wisdom and humanity of individual doctors.