番茄社区

CHICAGO -- When social media met personalized medicine, the result was longer survival for lung cancer patients, a researcher said here.

In a multi-center phase III randomized trial, patients using a Web application that triggered care based on reported symptoms did significantly better than patients getting standard care (hazard ratio 0.33, 95 % CI 0.16-0.67, P=0.0014), according to , of the Jean Bernard Cancer Center in Le Mans, France.

The efficacy results were sufficiently impressive at a planned interim analysis that the trial was stopped early, Denis told reporters at the American Society of Clinical Oncology annual meeting.

The advantage of the application, Denis said, is that it allows healthcare providers to react quickly to a change, rather than waiting for a scheduled visit. "Relapse doesn't occur during planned visits," he said.

"If [patients] were doing poorly, then they were immediately put in contact with their physician," commented , of the University of Colorado Cancer Center in Denver. "So they didn't get sick in the interval between physician visits."

The improved outcomes associated with the application underscore the need for close contact between patients and physician, Bunn, who was not involved in the study, told 番茄社区.

If the application were a pill, it would be big news, commented ASCO spokesperson , of the University of California Los Angeles, who moderated the press briefing "If we had a drug or an intervention that carried this level of survival benefit, wouldn't we want to go out and use it?"

The intervention, she noted, appears to be easy to use; is tailored to the patient; and allows care to be delivered when needed, rather than on a schedule -- something that should eliminate waste of healthcare resources.

The application, dubbed Moovcare, can be used on a computer or smartphone, Denis said.

Once a week, patients use the application to enter data on a suite of symptoms, including such things as asthenia, cough, dyspnea, and anorexia. A computer algorithm analyzes the information and, if warranted, triggers an email alert to the healthcare team, leading to a phone call or a visit.

In principle, use of the application should result in more appropriate and timely care, better control of symptoms, improved general health, and better cancer care, he explained.

To test the issue, the authors enrolled 133 patients and randomly assigned them to standard care or to care mediated by the Moovcare application. When the study was stopped, 121 patients were evaluable, 60 of them in the Moovcare arm.

Patients were at high risk, but were without progression at enrollment, and had a good performance status, with disease stages from II through IV but a relatively low symptom burden. In both arms, patients had planned appointments and tests on a standard schedule, every 3 or 6 months depending on their cancer stage. The primary endpoint of the study was overall survival (OS).

Of the 121 patients in the survival analysis, 90% were stage III or IV and they had a median age of 65.

After a median follow-up of 9 months, median OS was 19 months among those using the Moovcare app, and 11.8 months among those getting standard care.

Importantly, performance status at first relapse was 0 or 1 for 81.5% of the patients in the experimental arm, compared with 35.3% for those getting standard care (P<0.001).

Because imaging was done as needed rather than on a schedule, the experimental arm saw a 50% reduction in imaging when calculated per patient per year, Denis said.

Denis said he'd expect similar outcomes if doctors phoned every patient every week, but "it's not possible."

Lung cancer patients often delay reporting symptoms because they don't want to bother the healthcare team, commented ASCO spokesperson , of the Helen F. Graham Cancer Center in Wilmington, Del.

The important issue is to get a picture of symptoms and "this is one way to do that," Masters, who was not involved in the study, told 番茄社区. However, other approaches might also work.

Ganz noted that similar experiments have been tried in the U.S. and have shown that the key element is having someone -- usually not the physician -- in charge of responding to the data and getting the patient in for evaluation.

"When the feedback information has just come to the doctor, there's not much action -- this has been shown," Ganz told 番茄社区. So the main cost is not software or computers, she said, but having someone to monitor the system, she added.

Comment