Personalized Therapy Improves Response Rate, Survival in Phase I Study

Therapy directed at a particular molecular aberration in a patient's tumor resulted in higher response rates and extended time to treatment failure compared with standard of care in a phase I trial.

Although the study was not randomized and was limited by a number of factors, the findings are nonetheless important in highlighting the potential of personalized cancer therapy, according to Apostolia M. Tsimberidou, MD, PhD, of the University of Texas M. D. Anderson Cancer Center. Dr. Tsimberidou presented the results of the personalized medicine study (Abstract CRA2500) during Friday's Clinical Science Symposium, "Personalized Medicine."

During the study, if an aberration was identified through molecular analysis, that patient was matched with a therapeutic regimen designed to target that specific abnormality. A second group of patients in whom an aberration was not identified received therapy appropriate for their malignancy.

Patients treated under the matched protocol experienced significantly better complete and partial remission rates, as well as overall extended time to treatment failure and an improved survival rate (p < 0.0001). Dr. Tsimberidou explained that direct comparison between the two groups was limited by the inherent shortcomings of current molecular analysis technology and the infinite number of agents available to target tumor aberrations. Yet, she added, the results demonstrate that the concept of personalized medicine can be highligh beneficial.

"Discovering molecuar abnormalities — and selecting therapy particular to that molecular abnormality — is superior to the standard of care," Dr. Tsimberidou said.

Study Results

Of the 1,114 consecutive patients referred to M. D. Anderson for treatment under a phase I protocol, tissue samples were available for 955 patients; a molecular abnormality was identified in 852 of these patients. The investigators targeted genetic abnormalities known to affect tumor growth (Fig. 1). Of the 852 patients, 354 had more than one abberration. However, the analysis was limited to patients with one aberration, which totaled 175 patients, Dr. Tsimberidou said.



The 175 patients were treated in 36 different matched trials. “A therapy was considered to be matched to a patient’s tumor molecular abnormality if at least one drug in the regimen was known to target functionality of at least one of the patient’s molecular aberrations,” Dr. Tsimberidou said. There were 116 patients treated in 49 separate unmatched trials; no drug in their treatment regimen was known to inhibit their particular molecular aberration.

Total rate of complete or partial remission in the matched group was 27% compared with 5% in the unmatched group (p < 0.0001; Fig. 2). A greater percentage of patients in the matched group did not experience treatment failure, and patients in the matched group had a median survival of 13.4 months compared with 9.0 months in the unmatched group (p = 0.017).

Multivariate Analysis and Potential Confounding Factors

The study enrolled a number of patients with advanced or recurrent cancer. In a separate discussion of the study results, Paula M. Fracasso, MD, PhD, of the University of Virginia Cancer Center, suggested that aberrations in recurrent disease may be moleculary dissimilar. This, combined with the fact that patients underwent previous therapy (potentially affecting the outcomes of the matched or unmatched therapy), may have confounded the results, Dr. Fracasso said.

The analysis included only "a limited number of driver mutation," Dr. Fracasso said. The study investigators also did no perform subgroup analysis to identify treatment efficacy according to molecular abnormality. In other words, Dr. Fracasso said, there may have been significant differences in the type and number of mutations analyzed, as well as in the results generated, despite baseline characteristics being similar in the matched and unmatched therapy groups.

Despite these limitations, Dr. Fracasso said that it is time to “change the paradigm,” and begin thinking about directing therapy at a molecular target rather that developing organ-specific treatments.

Since the inception of the study, several new biomarkers and testing technologies have come to the market. In the future, Dr. Tsimberidou said, it will be important to conduct randomized trials to further elucidate the effect of personalized therapy.

M. D. Anderson is developing the capacity to perform genetic profiles of all new patients with cancer, which is approximately 30,000 new patients a year, Dr. Tsimberidou said. The program should be up and running in approximately 5 years, she noted.