The Food and Drug Administration last May ordered a major drug firm to stop selling its "new and unique" form of Acutrim, a popular nonprescription diet pill, and threatened to seize the product if the firm did not comply.

Six months later, after strenuous claims by the CIBA-GEIGY Corp. that its diet pill was not all that new, FDA backed off, saying, "We do not believe that regulatory action is indicated at the present time."

The reversal is now being questioned by House Government Operations subcommittee Chairman Ted Weiss (D-N.Y.), who sees it as an example of the FDA "failing to meet its responsibility of insuring the safety and effectiveness of all new drugs before they are marketed." The agency's actions, he said, "signaled to the drug industry that new-drug approval requirements could be ignored."

Weiss and his subcommittee staff, who have been probing the agency's regulation of new drugs, are concerned that what they see as lax FDA enforcement lets many manufacturers flout the law.

At the center of the problem is one of the longest-running regulatory merry-go-rounds in Washington -- the FDA's unsuccessful effort to catch up with a nearly 23-year-old law creating a two-class system of "old" drugs -- presumed to be safe and effective -- and "new" drugs requiring extensive premarketing tests. In 1962, Congress added a requirement to the 1938 Food, Drug and Cosmetic Act that companies seeking marketing approval for new drugs prove their products effective as well as safe.

Under the same law, however, the FDA allowed sales of drugs already on the market and gave dispensation to their new chemical relatives. The agency then began to review these "old" products to determine if they met the 1962 "effectiveness" standard.

Twenty-three years later, many of the reviews are unfinished. In the meantime, while publicly touting their products as "new," many companies tell the government that they really are variations of old drugs and thus not subject to the law's requirements.

The problems with the situation were dramatized last year by the deaths of some premature babies fed an unapproved prescription-vitamin preparation called E-Ferol.

FDA officials admitted to Weiss' subcommittee on intergovernmental relations and human resources last spring that they failed to move against E-Ferol because they thought -- incorrectly -- that it was similiar to other vitamin preparations scheduled for examination as part of the review.

The FDA later gave the subcommittee a list showing that about 5,000 prescription drugs are being marketed without formal approval and argued that most are old drugs that pose no safety problems. But the agency also announced plans to act against unapproved prescription drugs that come on the market in the future and said that companies must provide reports documenting any adverse reactions to any drug, regardless of its approval status.

A list of unapproved nonprescription drugs has not been provided -- an FDA official said there is no such list -- and the subcommittee's deadline of Jan. 16 passed without an FDA explanation of its decision to leave Acutrim on the market.

"Acutrim is one of perhaps thousands of new nonprescription drugs being sold without required agency approval," Weiss contends.

FDA spokesman William Grigg said recently that the Acutrim action was a "close call, not only in terms of regulatory and legal matters but scientific aspects as well." He defended FDA's reversal, saying that agency scientists eventually concluded that the product was similar to other time-release diet products on the market.

To pursue Acutrim would have been "selective enforcement on an arbitrary and what we thought irrational basis," said Jerome Halperin, who in 1983 moved from the deputy directorship at FDA's Bureau of Drugs to a CIBA Consumer Pharmaceuticals vice presidency. "How is the line to be drawn?" he asked.

The safety and effectiveness of nonprescription diet pills as a group are still awaiting action under the agency's over-the-counter drug review program, which began in 1972. Agency policy allows diet pills marketed before December 1975 or similar products coming on the market thereafter to be sold until final regulatory action is completed. But diet pills deemed "new" need premarketing approval.

Acutrim went on the market in September 1983 and CIBA estimates that it accounts for about 12 percent of the lucrative diet pill market. Its original label called it "the 16-hour precision-release appetite suppressant that works by osmosis . . . . The first osmotic tablet in the United States."

The time-release system uses the natural process of osmosis to cause the drug's active ingredient, phenylpropanolomine (PPA), which has been sold for decades, to diffuse slowly from the pill's core.

The checkered history of FDA's scrutiny of Acutrim is shown in agency documents:

* Shortly after the company began national marketing of the drug, FDA's Office of Compliance asked its Newark district office to conduct an inspection to determine whether Acutrim's "new" time-release system warranted regulatory action. An FDA investigator's memo of the Dec. 15, 1983, inspection quoted company officials as calling Acutrim "the first product to be marketed using this new technology."

* On Feb. 10, 1984, FDA officials met with CIBA representatives.

* Rudolf Apodaca, director of FDA's division of drug-labeling compliance, urged in an April 17, 1984, memo that the agency order Acutrim off the market. Apodaca concluded that "clearly the unique time-release dosage form causes the article to be a new drug." Failure to act against Acutrim, he warned, would "jeopardize the entire compliance policy."

* In early May, the FDA's general counsel's office agreed to move against the diet pill. On May 24 FDA's Daniel L. Michels, director of the Office of Compliance for drugs and biologics, sent CIBA a letter ordering that Acutrim be taken off the market.

* A June 6 letter from CIBA reiterated the compnay's earlier arguments that FDA's action "constitutes an arbitrary and discriminatory action." Although Acutrim's time-release mechanism was new, CIBA said, its dosage level and effect on the body were similiar to those of other diet pills on the market. The letter added that the seven top diet pills had been marketed after the 1975 deadline, without FDA interference.

* On June 13, FDA and CIBA officials met again.

* An Aug. 27, 1984, memo from FDA lawyer Kenneth Baumgartner to Michels in the compliance office reported that officials of the agency's Center for Drugs and Biologics had decided "to take no further action regarding Acutrim" and included a draft letter to the company. He noted that it was "short by design. In the regulatory letter we said that the time-release system was new. Since we are not taking further action we must give a logical reason -- it's really not so new -- without setting too much precedent or creating an administrative quagmire."

* In a four-paragraph letter Nov. 26, Michels told CIBA that no further action would be taken "at this time." He said "there appears to be no safety risk" from Acutrim.

The letter also said a 1982 policy exempting certain kinds of over-the-counter diet drugs from regulation was "arguably ambiguous."

The policy was based on an FDA-sponsored report that concluded that PPA was safe and effective at certain doses, but questioned whether higher levels might trigger potentially hazardous blood pressure increases.

FDA spokesman Grigg said recently that continuing studies must be completed before the agency develops an overall regulatory policy for nonprescription diet pills.