Samples of the controversial anti-cancer drug Laetrile seized by U.S. Customs agents have been found by several prestigious testing laboratories to be far below stated strength, chemically different from sample to sample and contaminated with fever-causing fungi.

The material, in liquid tablet form, was manufactured by Cyto Pharma de Mexico, the major producer of the Laetrile used by cancer patients in this country and at several large clinics in Mexico.

However, "all samples were determined to be chemically subpotent, mislabeled and of poor manufacturing quality. More than 20 samples of the ampuls (of liquid Laetrile) were found by visual inspection to contain . . . contamination," according to a report by the chief of the National Cancer Institute's Pharmaceutical Resources Branch.

One of the things that most disturbs J. Paul Davignon, the NCI pharmaceutical chief, is that the ampuls of liquid Laetrile, supposedly pure and intended for injection into the veins of humans, contain no lot numbers. Therefore, if one ampul proves to be contaminated, there is no way to know how many others are bad.

"I've never experienced this before," said Davignon, referring to the lack of lot numbers. "It's inconceivable that a manufacturer would produce a product without lot numbers. I question whether these people are in the pharmaceutical business."

Reached by telephone in Tijuana, Mexico, Cyto Pharma general manager George Del Rio said the product in question "belongs to somebody else. There's something wrong here. All our ampuls are lot-numbered.

"Every batch from this laboratory is certified by the Health Department of Mexico," said Del Rio. "All of my lots have to spend 40 days on observation. These may have been opened for two of three houre and then resealed."

The Laetrile tested for the National Cancer Institute by the Stanford Research Institute, Midwest Research Institute and several other laboratories was all seized at various times by customs inspectors at the Mexican-American border below San Diego, Calif.

The liquid Laetrile is contained in heat-sealed glass containers which cannot be opened without breaking them. The labels are baked-on paint, and do not appear to have been tampered with in any way.

While only 20 to the 2,715 ampuls were found to be contaminated, Davignon said "if you find one ampul that causes fever you would expect the entire batch to act the same way . . . the product was exposed to some organism."

The fungi were tested in rabbits, he said, and caused fever and general lethargy. "This is a bad product," said the NCT official.

In addition to the contamination, the scientists found that the Laetrile, which sells for about $7.50 per ampul, was only half as stong as its label indicated.

According to Davignon, the Laetrile was tested because National Cancer Institute was considering clinical, or human, trials of the controversial drug which the American Medical Association, the Food and Drug Association, the Sloan Kettering Institute and the American Cancer Society have all called worthless in the treatment of cancer.

The NCI was considering the trials, however, because more than a dozen state legislatures have legalized the use of the drug manafactured from the seeds of aprictos and other fruits rich in natural cyanide.

In a letter, to a researcher at the Mayo Clinic, in Rochester, Minn., Davignon summarized the test results and said. "There is no way that I could release these products in good conscience for human use."

Dr. Bayard H. Morrison, assistant director of the National Cancer Institute, said yesterday that NCI had never "committed ourselves to going ahead with human trials), we said we were condering it . . . It hasn't been abandoned, it's been postphoned."

Instead, officials of the NCI the FDA and the National Center for Disease Control have been meeting to organize a retrospective study of cancer patients who have already used, or are using, Laetrile.

"The idea would be to try to take advantage of the large group of people out there who have taken Laetrile in the past few years," said Morrison. In a sense, the experiment has already been conducted, but not in a controlled fashion.

"The idea is to get some information on these people who have been treated," said Morrison. "people on whom enough information has been compiled to get an idea of the effect of Laetrile on them."

Morrison said that the project would not simply involve asking Laetrile users to send in testimonials, which they have been free to give in the past, but rather researchers would try to gather data concerning the patient's type of cancer, treatment previous to Laetrile, and effects of all treatments.

If such a study showed that Laetrile appeared to have a beneficial effect, the National Cancer Institute might then reconsider testing the substance on cancer patients, said Morrison, who added that the Institute would have to find an uncontaminated supply of the drug.

According to Davignon, NCI is prepared to manufacture its own Laetrile. "When and if they make any decision like that (to begin tests), the groundwork will already be behind us," he said.