Nearly eight years ago, a unit of the global drug firm Allergan paid $150 million to settle criminal charges of illegally marketing three drugs. Now, Andrew Lelling, the US attorney in Boston, is being asked to reopen an investigation after newly unsealed documents suggest the company may have deliberately misled federal officials about a key clinical trial for one of the medicines, undermining the basis for the settlements.

The documents indicate Forest Laboratories — now part of Allergan — obscured crucial information demonstrating the Celexa antidepressant was not effective in children. Instead, using what its medical director acknowledged was “a masterful stroke of euphemism,’’ the company portrayed the study results as positive in materials that were submitted to the Food and Drug Administration in hopes of winning approval for pediatric use, according to the court documents.

It’s the same study that Forest, which was based in New York, had improperly touted as evidence its pill was useful in helping children, even though regulators had never approved Celexa for that purpose.

The request to reopen the investigation was made by lawyers representing consumers and pension funds, which claim they overpaid for medicine that was marketed illegally and later shown not to be effective.

In the lawyers’ view, Forest committed “misconduct and fraud’’ and may have warranted a different penalty had the negative MD-18 results been known. Forest pleaded guilty to one count of obstruction and two counts of distributing a misbranded drug under the Food, Drug, and Cosmetic Act. One of the counts was specifically related to promoting Celexa for use in children and teenagers between 1998 and 2002.

“Our litigation has revealed that the scope and extent of Forest’s fraud was not honestly disclosed to the US attorney’s office or to the Food and Drug Administration and that Forest misrepresented material facts underlying the prosecution,’’ they wrote in a Jan. 24 letter to Gregg Shapiro, an assistant US attorney in Boston.

The lawyers also argue that another Forest antidepressant called Lexapro might never have been approved for use in adolescents if the company had properly conveyed information about the negative study. In a deposition, a former FDA official indicated the agency relied on the results to later endorse Lexapro.

A spokeswoman for Lelling’s office in Boston would say only that “we don’t confirm or deny any possible investigations.’’ A spokeswoman for Allergan, which acquired Forest in 2014, declined to comment.

The episode began with a packaging error in 2000, when a few of the first nine patients enrolled in the clinical trial were inadvertently given pink Celexa pills instead of white placebo pills. The mistake was reported to the company by two investigators. In scientific terms, the trial investigators were unblinded, because they would have known which patients were getting which pill. This meant the results from those patients should not have been included in a final analysis presented to the FDA.

Nearly two years ago, a trio of researchers, two of whom worked as expert consultants for one of the law firms involved the class action litigation over Celexa and Lexapro, published a paper that discussed the packaging error and the positive outcome of the trial when excluding those patients.

What was not known until certain documents were unsealed by a federal court in Boston was the company reaction.

The mixup triggered substantial debate. Forest sent letters to all study investigators alerting them to the problem, according to the documents. A separate letter, meanwhile, was sent to the FDA to inform agency officials of the problem but also to give them a heads-up on what to expect when a final report on the study would be submitted.

At this point, the lawyers contend, the company began obscuring facts. In an initial draft of the letter sent to FDA, Forest acknowledged the incorrectly dispensed pills could have “potentially unblinded the study,’’ according to the unsealed documents.

In the final study report, the company reiterated the packaging error involving nine patients. Forest also acknowledged that the Celexa and placebo pills were “distinguishable in color, although otherwise blinded.’’

“Due to this error, medication was dispensed to eight patients in a fashion that had the potential to cause bias,’’ a Forest regulatory affairs manager wrote to the head of the FDA neuropharmacological drug products unit.

One expert on clinical trial transparency opined that the episode and the effort to push the feds to revisit the case offers an important lesson.

“These are all efforts that ultimately may get us closer to correcting the scientific record,’’ said Dr. Peter Doshi, an assistant professor of pharmaceutical health services research at the University of Maryland School of Pharmacy.

“If we want to trust in the process of medical decision-making based on evidence, then we need to make sure the evidence is trustworthy.’’

Ed Silverman can be reached at ed.silverman@statnews.com. Follow him on Twitter @Pharmalot. Follow Stat on Twitter: @statnews.