Judy Perkins, right, paddles during a team canoe race in Wilton Manors, Fla. Perkins has undergone an experimental breast cancer treatment and has seen successful results. (Photo for The Washington Post by Scott McIntyre)

A Florida woman diagnosed with advanced breast cancer, generally considered incurable, is free of the disease two-and-a-half years after a novel therapy used her own immune cells to target her tumors, researchers said Monday. Striking recoveries were reported earlier for a patient with deadly liver cancer and another with advanced colon cancer.

The three patients were treated by a team at the National Cancer Institute led by Steven Rosenberg, an immunotherapy pioneer who is chief of the surgery branch. For each patient, the team sequenced the genomes of their tumors to find mutations, then tested immune cells extracted from the cancers to identify which ones might recognize the defects. Those cells were expanded by the billions in the laboratory, then infused back into the patients, where they attacked the tumors.

Rosenberg stressed that the approach, called adoptive cell therapy, is experimental and that several other patients who got the same treatments had not responded. But he said the three cases point to a potential “blueprint” for targeting a wide range of advanced solid tumors of internal organs, including the stomach, esophagus and ovary. Such malignancies account for the vast majority of the 600,000 cancer deaths that will occur in the United States this year.

The approach relies on mutations, not cancer type, he said. “The very mutations causing the cancer will be the Achilles’ heel.”

The breast-cancer case appeared Monday in Nature Medicine. The other cases were described in scientific journals in 2014 and 2016.

The latest case involves Judy Perkins, a 52-year-old structural engineer who lives in Port St. Lucie, Florida, and had a mastectomy after being diagnosed with very early stage breast cancer in 2003. A decade later, she learned cancer had spread to other parts of her body; she underwent multiple treatments, all of which ultimately failed. A chance meeting with a top NCI scientist prompted her to enroll in the trial in 2015.

One of her tumors, removed surgically, had 62 different mutations. Researchers also removed immune cells from the malignant mass – tumor-infiltrating lymphocytes, or TILs – and found some that targeted four of those defects. They then expanded their number into the tens of billions. Perkins first got chemotherapy, followed by the infusion of immune cells.

Five months later, her scans were clear. The cancer has not returned.

“My TILs army,” she calls it, admitting she never expected the outcome. “I’m blown away.”

Experiments in one patient, or even a small number, don’t prove that a treatment will be effective in others. Rosenberg’s team has treated more than 40 patients with common solid tumors over the last four years using the highly personalized TILs therapy; most entered the trial with very limited life expectancy, and about 15 percent have responded in some way.

But those responses vary, said Stephanie Goff, a clinician on the team. Perkins, as a “complete responder” who has needed no further treatment, is the exception. The researchers are “trying to find ways” to achieve more consistent outcomes, Goff said.

Still, the recoveries in the three individual cases now reported represent an important advance, according to the researchers. While immune therapy has benefited patients with advanced melanoma, lung and some other malignancies with large numbers of mutations, it generally has not been effective against cancers that start in the linings of organs – called epithelial cancers – and have fewer mutations.

Other scientists welcomed the news but were cautious. Carl June, an immunotherapy expert at the University of Pennsylvania, said the breast-cancer patient’s response to TILs was “striking.” Yet was this case “one in a million,” he asked, “or something that is an approach that will benefit many women?”

Scott Antonia, an immunologist at Moffitt Cancer Center in Tampa, Florida, said further work is crucial. “The paper shows the biological proof of principle and that you can use this approach to have a profound impact on a patient,” he noted. “We should continue to go down that path.”

Melinda Bachini, a former paramedic who lives in Billings, Montana, believes the treatment saved her life. In 2009, she was diagnosed with bile-duct cancer that had spread to her liver. She was 41.

Bachini’s initial surgery removed two-thirds of her liver, but three months later the disease showed up in her lungs. She had several treatments that weren’t successful, then found the NCI trial online. In 2012, she got her first TILs infusion, and her tumors began to shrink.

When they again began growing, she received another batch of cells – more aggressively targeted to her mutation – the following year. The tumors shrank again. In fall 2016, she needed more treatment and returned to NCI for a different kind of immunotherapy. Today, she said, she has a “few spots” in her left lung that Rosenberg believes may be scar tissue.

The patient with advanced colon cancer whom Rosenberg’s team treated in 2015 is Celine Ryan of Michigan. Most of her tumors disappeared after she got TILs, though one in a lung worsened and required more surgery. She remains free of cancer today, Rosenberg said – the first successful targeting of a deadly mutation called KRAS, which is implicated not only in some colon cancers but in pancreatic and lung cancers.

Perkins remembers her oncologist giving her three years to live when her breast cancer returned in an advanced form in 2013. While trying treatment after treatment, she became a breast cancer advocate and went to California for training by Project Lead, a program run by the National Breast Cancer Coalition. The lecturer on immunology was Goff, on Rosenberg’s team, who mentioned the trial. Perkins said she “cornered her with a beer” and said she wanted to enroll.

Perkins hopes she’s cured – a word she is almost afraid to utter – but knows that the cancer “could come back tomorrow.” She’s very aware that most patients with metastatic breast cancer aren’t as fortunate and that experimental approaches can entail big risks.

A South Carolina woman named Janice Satterfield contacted Perkins after reading a post she wrote about the NCI trial. Satterfield ended up enrolling in the trial and in August 2016 went to NCI for treatment. She developed complications and died a few months later.

“She may have lost her life,” her husband, Scott Satterfield, said recently, “but I hope someone else will gain from her trial.”