As oncologists race forward with new treatments verging on science fiction and biotech companies press on with drugs for once-hopeless rare disorders, one of the world’s most pervasive diseases looks as it’s been left behind.

There are few new drugs on the horizon for diabetes, which affects about 29 million Americans. Most of the treatments in late-stage development are simply improved versions of what’s out there — taken weekly versus daily, or orally instead of by injection.

So has pharma run out of ideas on diabetes?

Not exactly. But whether its ideas will ever get to market is another question. There’s plenty of promising science in the early stages of research. Available drugs, however, work pretty well. Given the cost of development and a high bar for approval, pharma can only afford to advance true-blue breakthroughs, and those are hard to come by.

“There are unmet needs, but it’s going to take a really good drug,’’ said Dr. John Buse, director of the Diabetes Center at the University of North Carolina Chapel Hill. “It’s not like 10 years ago where you could market acarbose, a drug you have to take three times a day that makes you fart. Now that profile would never make it into Phase 2.’’

And the economics can be unforgiving. Oral insulin has long been considered a Holy Grail in diabetes research, potentially freeing patients from routine injections in favor of an easy-to-take pill. But late last year, the Danish drug maker Novo Nordisk backed away from such a project — not because it failed in clinical trials, but because developing it would have been too expensive to get a sure return on investment.

That logic frustrates patients and their advocates.

“My wish would be that people thinking like that would spend time with the people who face the challenges of Type 1 diabetes,’’ said Derek Rapp, chief executive of JDRF, which funds and advocates for research in the field.

But diabetes has become a tough and crowded market. Industry leaders Sanofi, Novo Nordisk, and Eli Lilly have all issued grim sales forecasts in recent years, as pricing pressure makes their past discoveries less and less lucrative. Novo Nordisk’s CEO abruptly decided to resign last year amid a surprise decline in projected revenue.

“Clearly the bar has gone up,’’ said Dr. Philip Larsen, Sanofi’s head of diabetes research. “The larger companies are now in what you could call a thinking-outside-the box mode.’’

That means betting on early-stage efforts that could truly move the needle, albeit with high odds of failure.

Scientists at Sanofi, for instance, noted that gastric bypass surgery can have remarkable effects on Type 2 diabetes, even sending some patients into remission. Their question: Can that be replicated with a drug? They’ve got a drug, now in mid-stage development, that seeks to “copy the cocktail of endocrine events that happens in the aftermath of surgery,’’ Larsen said. The goal: Create an injection that spares the need for a knife.

Drug makers are also at work on so-called smart insulin, which would switch on when blood glucose is too high and then harmlessly switch off once it had normalized, reducing the risk of hypoglycemia. Others are digging into the gut microbiome, a network of trillions of microbes, in hopes of finding a way to ameliorate diabetes.

“We don’t want to add new classes just to have new classes,’’ said Dr. Ruth Gimeno, vice president of diabetes research at Eli Lilly. “We want to really make a difference. We’re giving ourselves a little more time rather than saying let’s just fill the pipeline with incremental things.’’

And pharma has gradually widened its aperture to include novel technology, examining how wearable devices, glucose monitors, and advanced insulin pumps can work alongside drugs to improve patients’ lives.

“There are some exciting and promising things that are being looked at, but I think the scope of diabetes management is getting a bit wider to be not strictly pharmaceuticals,’’ said Dr. Robert Gabbay, chief medical officer of Harvard University’s Joslin Diabetes Center. “Devices are becoming increasingly important.’’

For some doctors, the key isn’t more new products. It’s education — to teach them how to use what’s already on the market

Drugs called GLP-1 agonists, introduced more than a decade ago, have been proven both to lower blood sugar and cut the risk of long-term cardiovascular problems. The same goes for SGLT2 inhibitors, introduced in 2013, which have been shown to reduce the risk of death by 32 percent for Type 2 diabetics.

And yet they’re not being widely used.

That’s in part because most primary care physicians don’t know how to use them, said Dr. Athena Philis-Tsimikas, who leads the Scripps Whittier Diabetes Institute in California. The other issue is paperwork: Doctors have to go through “a ridiculous number of steps’’ to get patients’ health plans to cover those new drugs, Philis-Tsimikas said.

“It’s not like we’re desperate for something that is additionally innovative,’’ she said. “We’re not using what we have.’’

Damian Garde can be reached at damian.garde@ statnews.com. Follow him on Twitter @damiangarde. Follow Stat on Twitter @statnews.