Novartis (NVS - Get Report) in 2015 was required to divest two melanoma treatments it was developing in order to secure Federal Trade Commission approval for the company's $16 billion acquisition of GlaxoSmithKline's (GSK - Get Report) cancer-treatment portfolio.

Novartis' two melanoma treatments, known as BRAF and MEK inhibitors, were also being developed to treat a variety of other cancers. The market for the two inhibitors was concentrated—Novartis and GSK were two of a small number of companies with either a BRAF or MEK inhibitor on the market or in development, and were two of only three companies marketing or developing a BRAF/MEK combination product to treat melanoma.

The Novartis products, sold to Array BioPharma (ARRY) , were in final trials for a combination product to treat melanoma but also were in various stages of development for other types of cancer. The FTC argued that unless it required Novartis to divest its BRAF and MEK inhibitors, Novartis would likely delay or terminate their development as well as the combination product, ultimately raising prices for cancer treatment and potentially depriving consumers of break-through products.

The episode highlights the concern the FTC has about pharma mergers that could consolidate two drug development pipelines into the hands of one company when the merging firms are working on potentially competing products. The regulators have two main worries when that happens. One is that under a single owner, two possible rivals won't compete with each other anymore and consumers won't enjoy the lower prices that competition brings. The other worry is that the buyer will devote resources to only one of the products and the other will be left on the shelf. If the first one fails to win approval from the Food and Drug Administration, neither drug candidate will make it to market.

But forced sales like the one Novartis made can also deplete a company's stable of drug compounds, leaving them with few fall-backs if their lead prospect fails.

The harm to a pharma firm's pipeline was on the mind of Novartis' main U.S. antitrust lawyer when he recently called for the FTC to stop requiring drug pipeline divestitures unless the products were in the final round of FDA testing, known as Phase 3. Clinical trials during Phase 3 are meant to demonstrate whether a product offers the health benefits the company intends and what the long-term side effects may be. Phase 1 and Phase 2 focus on safety and efficacy of a drug. After successful Phase 3 results a company may submit an application to market the drug to the FDA.

"If you're going to consider potential competition, I don't see how anything before Phase 3 can meet the standard under the case law because you're talking about a low likelihood of getting to market," David Emanuelson, Novartis senior corporate counsel for antitrust in the Americas said last month at the American Bar Association's spring antitrust meeting.

"Phase 3 is where you really are investing significant money to run a controlled trial against a placebo ... to see if it's better than the [current] standard of care."

He noted that statistics tracing clinical trial results show that products in Phase 1 or 2 typically have a 10% to 15% likelihood of winning FDA approval. At Phase 3, the probability jumps to 50%.

Emanuelson complained that antitrust enforcers' new guidelines on investigating mergers for harm to innovation competition removed a safe harbor that made allowances for the number of competitors. "You have to take a look at who else is doing research in the area. It has to be an element of the competitive analysis," he said. "You can't just rely on the fact that companies are doing competing development and after a merger one might consolidate the development into a single program. Because if 10 other companies are engaged in that same R&D, that's an efficiency, not an anticompetitive effect." (Under antitrust law, regulators should consider whether economic efficiencies caused by a merger will offset whatever consumer harm might be caused by the increase in concentration.)

"In a costly market like pharma development, you want to see efficiencies created and companies getting more production out of the capital they invest in new development."

But FTC Competition Bureau assistant director Michael Moiseyev, when asked by The Deal about Emanuelson's comments, said the commission does consider the probability that a drug will make it to market among other factors it considers when reviewing drug pipeline mergers but there should not be a hard rule stipulating that only drugs in the latest stages of development be slated for divestiture.

"I understand the sentiment that the prospect of harm to competition from drugs in pre-Phase 3 trials can be remote, but these kinds of bright-line rules haven't served antitrust enforcement very well," he said. "There is a huge volume of commerce in pre-Phase 3 drugs. It's in the billions and billions of dollars in IPOs, venture funding, company resources and M&A activity. It's not like we're talking about the tail of the whole industry. Clearly investment in these products make sense because there's real money being put behind them. It doesn't seem like the business has any problem making huge bets on early-phase drugs."

The probabilities that early-phase drugs will make it through the FDA approval process is actually "all over the chart," he said. "Some have a high probability of making it to market, notwithstanding that they're in an early phase of investigation. They have a proven method of action, a proven physiological target or are follow-on drugs in a category that already has proven successful. That belies the proposition that there is uniformly a low probability of success in earlier phases of development."

Moiseyev added that given the size of many of drug markets, even "a somewhat remote, less than 100% certainty of competition can still have very significant competitive impact."

But Michael Lawhead, a shareholder in Stradling Yocca Carlson & Rauth, P.C.'s corporate and securities department and a member of the firm's life science practice group, said that the harm to innovation from consolidation of early stage drug pipelines is easy to overstate.

"For drugs in Phase 3, the FTC [would contend] the buyer already has products in the market and the target doesn't yet," Lawhead said. "In perceiving harm to innovation competition, the thinking would be that the buyer has diminished incentive to continue develop the target's pipeline compound because the product might divert sales from its own. In Phase 2 you don't have that risk but the FTC still may require you to divest assets. Many of companies will have to go through the cost and time of figuring out who will be the buyers for the assets."

He suggested abiding by Emanuelson's suggestion for enforcers to ignore early stage pipeline overlaps. "It's premature to require parties in a merger to divest products in Phase 1 or 2," he said, simply because they're unlikely to stop funding them at that point, and a buyer may be just as likely to do so. "There's no guarantee the buyer will be able develop them, so you're not addressing the competitive harm of the deal by putting those products in someone else's hands."

Editors' pick: Originally published April 28.