The National Cancer Institute failed for almost a year to tell doctors, patients or the Food and Drug Administration that an anticancer drug they are testing on thousands of cancer patients can cause kidney failure, FDA officials charged yesterday.

Investigators have found numerous cases of severe kidney damage in patients, including children, who have taken the drug, which is known as methyl CCNU.

That and other allegations of violations of FDA regulations for testing new drugs by the NCI, were made yesterday by Stuart Nightingale and Alan Lisook of the FDA before a hearing of the House science subcommittee on investigations.

If the charges are confirmed, the FDA has the power to halt all further trials of the cancer drug and can bar researchers from ever conducting another drug investigation for the government. In this case a letter containing the allegations and asking for corrective action will be sent to the NCI as an agency and, after NCI replies, the FDA will decide on its action.

The Washington Post has learned that the FDA testimony was based on a series of documents that the FDA found in the files of the cancer institute. They include one letter to the institute dated Nov. 13, 1978, four months before NCI notified anyone of the danger, in which Dr. Harvey Cohen of the Children's Hospital of Boston described at least 14 cases of kidney damage that occurred in cases where the drug had been used.

Lisook said the FDA received an angry letter in September, 1978, from a parent whose child's brain tumor was treated with the anticancer drug. The mother told the FDA that she had been informed of other possible side effects of the drug, which is used to treat brain tumors, but not of the possibility that her child might suffer kidney failure.

An NCI spokesman acknowledged that the institute learned of that case and two others in which the anticancer drug and kidney failure might be connected in June, 1978, and immediately started an investigation that took six months. After the investigation confirmed the connection between the drug and the kidney failure, the NCI notified FDA, the institute spokesman said.

FDA regulations require that cases of serious side effects be reported within hours or days of the time they are discovered.

In this instance, The FDA officials said that the time lag between the NCI's learning of the kidney problems caused by the drug and its report was at least 10 months, possibly longer.

When FDA went to the NCI to investigate the matter, in July, 1979, Lisook said, "we were quite surprised to find that they [the NCI] couldn't identify who the clinical investigator was, they couldn't identify what the protocol was . . . and they could only speculate why the [doctors giving out the drug] were not notified" about the problem of kidney failure."

The revelation of the FDA charges came out in the second day of hearings looking into scientific fraud and misbehavior, during which the repeated allegation was made that science was little or no apparatus to handle faked data, plagiarism and other scientific crimes.