Three patients with advanced cancer suffered no serious side effects from being treated at the University of Pennsylvania in the first U.S. clinical study of cells edited with CRISPR, the gene-editing technology.

But neither did they benefit, according to results published last week in Science. One patient with a bone marrow cancer called multiple myeloma has died and another has progressed. A patient with sarcoma, a soft-tissue cancer, also progressed.

Pilot clinical trials are designed to assess safety, not effectiveness. And the inaugural U.S. test of CRISPR-edited cells in humans was so ethically and scientifically fraught that Penn spent more than two years getting necessary approvals for the January 2018 launch.

Still, the experiment was intended to combine and improve on two revolutionary immune-boosting approaches that showed startling effectiveness even in early trials. One approach, which cuts a natural brake on the immune system, has led to a class of cancer drugs called checkpoint inhibitors. The other approach genetically engineers immune soldiers called T cells to recognize and attack cancer cells; the first approved T-cell therapy, Novartis’ Kymriah, was pioneered at Penn and Children’s Hospital of Philadelphia.

» READ MORE: The face of a medical revolution marks five years cancer-free — and her 12th birthday

Although the CRISPR-edited cells did not melt away tumors or stop cancer progression, the cells did survive and grow in patients for up to nine months. In a previous Penn clinical trial that used engineered T cells to attack multiple myeloma, half the cells were dead in a week.

Penn T-cell researcher Carl June, principal leader of the new study and previous groundbreaking work, sees the CRISPR trial as another incremental step in a medical odyssey. CRISPR technology was invented just eight years ago, yet the version used by Penn is already so outmoded that the trial has been discontinued.

"We learned what we wanted to learn,” June said of the study, on which Stanford University collaborated. “It opens up the door for a lot of new approaches.”

What they learned was mostly reassuring.

The difficult, many-step manufacturing process was feasible. CRISPR was used on T cells to cut out two genes, one that codes for the immune-system brake, and another that could hamper the T cells’ ability to latch onto cancer cells. CRISPR also inserted a gene that enabled the T cells to recognize and target a protein found on the cancer cells but not healthy cells.

These edits accidentally caused some unusual rearrangements of DNA in a small fraction of T cells — one of the biggest worries with CRISPR. But as the T cells took hold and multiplied in patients, these rearrangements steadily decreased, “suggesting that they conferred no growth advantage,” the researchers wrote.

The edited T cells did not trigger any of the dangerous toxicities related to revving up the immune system that are common with checkpoint inhibitors and engineered T cells.

However, the researchers noted that a longer trial with more patients and higher doses will be needed “to fully assess the safety of this approach.”

A problem that has occurred with engineered T-cell therapies also showed up in the CRISPR trial: one patient’s cancer cells stopped making the protein, called NY-ESO-1, that the T cells had been edited to target.

In future trials, “we would not just use NY-ESO-1 because we’ve learned the tumor can live without it,” June said. “We’d want to use multiple targets.”

Renier J. Brentjens, an oncologist and cell-therapy researcher at Memorial Sloan-Kettering Cancer Center in New York City, called the paper an important “proof of principle.”

“The T cells persisted and found the tumors. It would have been nice to see remissions or tumor regression, but it doesn’t necessarily mean the approach is flawed," Brentjens said. “It may be that the target they used is not sufficient."