Apostolia-Maria Tsimberidou MD, PhD, discusses the findings from the IMPACT trial and how precision medicine can affect patient care going forward.
Apostolia-Maria Tsimberidou, MD, PhD
The use of targeted matched therapy more than doubled 3-year overall survival (OS) in patients with refractory cancers compared with those who received nonmatched therapy, according to results from the IMPACT trial.
Lead author Apostolia-Maria Tsimberidou, MD, PhD, professor in the Department of Investigational Cancer Therapeutics at The University of Texas MD Anderson Cancer Center, presented these results at the 2018 ASCO Annual Meeting. She said that findings from studies like IMPACT exploring precision medicine could play a major role in cancer care, especially for patients with rare or difficult-to-treat diseases.
“We hope that with implementation of precision medicine through ongoing clinical trials such as our IMPACT2 randomized precision medicine trial, which is the next step, as well as, more importantly, ASCO's TAPUR study, we help improve clinical outcomes for patients with cancer,” she said. “I am optimistic that with implementation of precision medicine, we will be able to improve patient outcomes and, hopefully, one day, cure cancer using tumor testing and the best drugs possible.”
IMPACT was designed to evaluate outcomes for patients with rare cancers or who had exhausted their treatment options. From 2007 to 2013, investigators conducted molecular testing on 3743 patients and found at least 1 molecular alteration in 1307 (35%). Tsimberidou said 1 to 50 genes were analyzed and included genes in the PI3K/AKT/mTOR, MEK/RAF, and RET pathways.
The patient population mostly included those with gastrointestinal cancers (24.2%), gynecological cancers (19.4%), breast cancer (13.5%), melanoma (11.9%), and lung cancer (8.7%). Patients were then assigned to treatment designed to block the function of the mutated or altered gene (54.4%) or unmatched treatment (45.6%). Matched treatments could include targeted therapy in combination or alone.
In a retrospective analysis of data from consecutive, prospective molecularly profiled patients, the 3-year OS rate was 15% for those assigned to matched targeted therapy compared with 7% for patients who did not receive precision therapy.
The median OS (9.3 vs 7.3 months; HR, 0.72; P <.001) and 10-year OS rate (6% vs 1%) also favored the matched group. The matched group also had superior progression-free survival (PFS; 4.0 vs 2.8 months; HR, 0.67; P <.001). The objective response rate (ORR) was 16.2% with matched therapy versus 5.4% for nonmatched. Patients in the matched group were also more likely to have stable disease lasting ≥6 months, 18.7% versus 14.7%.
In an interview with OncLive, Tsimberidou discussed the findings from IMPACT and how precision medicine can affect patient care going forward.Tsimberidou: In 2007, we started the Initiative for Molecular Profiling and Advanced Cancer Therapy trial, or IMPACT, at The University of Texas MD Anderson Cancer Center. The rationale for this study was the dramatic improvement in OS that we noticed in patients with chronic myeloid leukemia after the introduction of imatinib (Gleevec).
The hypothesis of this study was that tumor molecular profiling would enable us to select the optimal therapy for patients with advanced metastatic cancer who were referred to our program for treatment. Until that time, the selection of treatment was random. After we initiated the IMPACT study, we ordered tumor molecular profiling for patients with advanced metastatic cancer. These patients had exhausted standard treatment options or they had rare, incurable tumors.
From 2007 until 2013, we ordered tumor molecular profiling. Tumor profiling included 1 to 50 genes, depending on the time of testing. In the first few years, we could only test for 1 or 2 genes using polymerase chain reaction. Towards 2013, we used next-generation sequencing that included a 50-gene panel in most cases.
We selected treatment in the clinical trial based on the results of patients' tumor molecular profile. If there was a clinical trial with a drug that was known to inhibit the function of patients' tumor abnormality, we offered the clinical trials with matched targeted therapy. If these trials were unavailable, patients were treated with nonmatched therapy.Patients were heavily pretreated—they had a median of 4 prior therapies ranging from 0 to 16. The median age was 57 years and 38% were men. We treated patients with advanced metastatic cancer of any tumor type including gastrointestinal tumors such as colorectal cancer, gynecologic tumors, breast, melanoma, lung, and other cancers.
The ORR was higher in the matched targeted therapy group compared with the nonmatched therapy group. The disease control rate, which included objective responses plus disease stabilization for at least 6 months, was 35% in the matched therapy group compared with 20% in the nonmatched group.
This improved disease control was associated with longer PFS. The median PFS was 4 months in the matched therapy group compared with 2.8 months in the nonmatched group. The hazard ratio was 0.67.
This improved PFS was also associated with longer OS. The median OS was 9.3 months in the matched targeted group compared with 7.3 months in the nonmatched therapy group. The HR was 0.72. The 3-year survival rate was 15% in the matched group compared with 7% in the nonmatched group. The 10-year survival rate was 6% in the matched therapy group compared with 1% in the nonmatched group.We also performed univariate and multivariate analyses to determine the association between baseline characteristics and OS. We found that PI3K/Akt/mTOR pathway abnormalities were an independent factor predicting shorter survival. Other factors were increased lactate dehydrogenase levels, lower albumin levels, performance status >1, liver metastases, increased platelet count, and age 60 years or older.
We developed a prognostic score using these baseline characteristics. We gave 1 point to each factor and we can now predict survival based on how many of these factors patients have. The median OS for patients with 0 factors was 15 months compared with less than 2 months for patients who had a score of 5 or higher.
Later, we added the type of therapy into the multivariate model. We found that matched targeted therapy was an independent factor predicting longer OS.
Therefore, our data demonstrate that matched targeted therapy is associated with a superior rate of response, PFS, and OS. We have demonstrated long-term OS data in patients treated with matched targeted therapy that were superior to those of patients treated with nonmatched therapy. We found that matched targeted therapy was an independent factor predicting longer survival. We found that PI3K/Akt/mTOR pathway abnormalities are associated with inferior outcomes and shorter survival.Since we started the precision medicine program at The University of Texas MD Anderson Cancer Center in 2007 with our IMPACT trial, the definition of precision medicine has evolved. Today, precision medicine refers to the use of targeted therapy to inhibit tumor abnormalities that are responsible for pathogenesis. However, it also refers to immunotherapy or other therapeutic strategies that are known to or will, we hope, inhibit carcinogenesis.
To implement precision medicine, in my opinion, you must have a complete understanding of tumor biology for each individual patient, including next-generation sequencing, tumor markers, immune markers, and cell-free DNA analysis, plus any other biologic data that are available using the most advanced technology. We have to have effective drugs that are known to inhibit the function of these biological abnormalities, using the rigorous definition of precision medicine. More importantly, we have to use effective drugs and conduct tumor testing at the time of diagnosis and during the course of disease to optimize treatment selection.
Tsimberidou AM, Hong DS, Wheler JS, et al. Precision medicine: clinical outcomes including long-term survival according to the pathway targeted and treatment period—The IMPACT study. J Clin Oncol. 2018;36(suppl; abstr LBA2553).