Matching treatment to specific genetic changes improves survival across multiple tumor types
The IMPACT (Initiative for Molecular Profiling and Advanced Cancer Therapy) clinical trial began in 2007 to evaluate the impact of personalized therapy selection based on molecular testing of tumours for people with difficult-to-treat cancers. During ASCO 2018, results of this study were presented indicating that matched targeted therapy results in a significant improvement in both progression-free (PFS) and overall survival (OS). Of the 1,307 patients with at least one genetic change in the tumour, the 3-year OS rate was 15% in the matched targeted group compared to 7% in the non-matched group. The 10-year OS rate was 6% for the matched group and 1% for the non-matched group. These findings show that molecular testing of tumours using next generation sequencing can be used to optimize therapy.
Patients referred to the study had advanced cancer that worsened despite standard treatment. The patient population was heavily pre-treated, with a median number of 4 prior therapy lines (some received 16 prior therapies). The most common types of cancer included gastrointestinal cancer, gynaecologic cancer, breast cancer, melanoma, lung cancer, and thyroid cancer. In the early years of the study, tumours were tested for mutations in individual genes, whereas in the late years of the study (until the end of 2013), next generation sequencing was used to test 20-50 different genes at once. Of the 3,743 patients who had molecular testing, at least one genetic change was found in 1,307 tumours. Of those, 711 patients received treatment matched to the biology of the tumour and 596 received a therapy that was not matched to the genetic change found in the tumour (because no matched treatment was available to the patient at the time). Matched therapies included single agent targeted therapy, as well as matched targeted therapy in combination with chemotherapy, or other treatments.
The objective response rate in the matched therapy cohort was 16.2%, while this was only 5.4% in the non-matched group. In addition to this, respectively 18.6% and 14.7% of patients in the matched and non-matched cohort experienced disease stabilisation. This corresponds to a clinical benefit rate of 34.9% in the matched therapy group. This was significantly higher than the 20.1% seen in the non-matched cohort (p< 0.001). The median PFS in the matched therapy patients was 4.0 months, while this was only 2.8 months in the non-matched group (HR: 0.67; p< 0.001). The OS was significantly better in patients receiving matched therapy as well, compared to non-matched patients (median OS 9.3 vs. 7.3 months; HR: 0.72; p< 0.001). Interestingly, the OS in the matched targeted therapy group plateaued, starting at 3.2 years (11% were still alive). The 3-year OS rate was 15% in the matched therapy group compared to 7% in the non-matched group. The 10-year OS rate was 6% for the matched therapy group and 1% in the non-matched group. Matched targeted therapy was found to be an independent factor predicting longer OS in multivariate analysis.
In a next step, the investigators developed a prognostic score to predict the OS of patients based on their baseline characteristics. Taking into consideration baseline characteristics in 1,307 patients, molecular alterations in the PI3K/AKT/mTOR pathway were an independent factor predicting shorter OS compared to other alterations. Other independent factors predicting shorter survival were the presence of liver metastases, elevated lactate dehydrogenase levels, poor functional status, low albumin levels, elevated platelet counts, and age ≥ 60 years.
In summary, outcomes were superior in patients who received a therapy that was matched to their specific genetic alteration. In addition to this, the study identified several factors that were predictive for a longer OS. In the matched therapy group, 10.4% of patients had an OS of more 38 months, or longer (the plateau starting point). As such, molecular testing of tumours using next generation sequencing can be used to optimize therapy and should be taken into consideration when selecting therapy for patients with difficult-to-treat cancers.
AM Tsimberidou, Hong DS, Wheler JJ, et al. Precision Medicine: Clinical Outcomes including Long-Term Survival According to the Pathway Targeted and Treatment Period: The IMPACT Study. Presented at ASCO 2018; Abstract LBA2553.