Real-World Data Underscore Importance of NTRK Gene Fusions in Tumor Agnostic Therapies

Oncology Live®Vol. 23/No. 13
Volume 23
Issue 13

The role of NTRK gene fusions across multiple cancer types have led investigators to confirm the marker in real-world data analyses.

Carsten Bokemeyer, MD

Carsten Bokemeyer, MD

The role of NTRK gene fusions across multiple cancer types have led investigators to confirm the marker in real-world data analyses.1

TRK fusion has become quite famous over the past few years, and we know this is a chromosomal rearrangement that drives oncogenesis in pediatric and adult cancers,” Carsten Bokemeyer, MD, said in a recent presentation of findings from a matching-adjusted indirect comparison (MAIC) of real-world data. “And there are specific inhibitors, such as larotrectinib, that selectively and potently block the activity from this fusion protein and have a high activity in patients with cancer.”

Larotrectinib (Vitrakvi) is a kinase inhibitor indicated for the treatment of adult and pediatric patients with solid tumors with an NTRK gene fusion without a known acquired resistance mutation, are metastatic or where surgical resection is likely to result in severe morbidity, or for whom no satisfactory alternative treatments exist or have disease progression following treatment.2

In an analysis presented at the 2022 American Society of Clinical Oncology Annual Meeting, larotrectinib demonstrated longer overall survival (OS) compared with standard-of-care (SOC) therapy in patients with TRK fusionpositive cancer.

MAIC is a common methodology that balances patient population characteristics allowing for cross-study comparisons in the absence of individual patient data, Bokemeyer said. The need for MAIC to assess the efficacy of larotrectinib in this patient population is because neurotrophic-TRK (NTRK) fusions are rare in patients, meaning that single-arm trials conducted to evaluate larotrectinib are limited in comparing its effectiveness with SOC regimens.

In their MAIC OS analysis, researchers found a 78% lower risk of death for patients treated with larotrectinib (n = 85) compared with SOC (n = 28), with a hazard ratio of 0.09 before weighting, but after adjusting it was 0.22 (95% CI, 0.09-0.52; P = .001). Moreover, after weighting, the median OS of patients on larotrectinib was 39.7 months compared with 10.2 months for patients on SOC treatment.

“We would need further data to confirm this, but this is one of the several analyses now performed in recent years that helps to add some information on how much the benefit of using NTRK fusion inhibition in this specific patient population really is,” said Bokemeyer, who is director of the Division of Oncology, Hematology, and Bone Marrow Transplantation at the University Hospital Hamburg-Eppendorf in Hamburg, Germany.

Investigators developed the 2 study cohorts leveraging insights from a previous analysis from the European Society of Medical Oncology that identified NTRK fusion–positive patients who did not receive TRK inhibitors and used this as the SOC cohort by using data from the Flatiron Health/Foundation Medicine clinico-genomic database.3 Then, researchers employed MAIC to compare larotrectinib with SOC by balancing various population characteristics among several studies. Patient-level data was collected from clinical trials with the TRK inhibitor: a phase 1 trial (NCT02122913), the SCOUT trial (NCT02637687), and the NAVIGATE trial (NCT02576431). They then aggregated SOC data by matching individual data on common baseline characteristics.1

A log-rank test of equality, to assess if the 2 groups were similar before the initiation of larotrectinib, found no difference between them (P = .31). After this, an estimation of the treatment effect of larotrectinib vs SOC on OS was conducted.

To assess the OS results with more certainty, researchers also conducted 2 sensitivity analyses as excessive matching variables in MAIC can reduce the effective sample size (ESS) and the larotrectinib follow-up was longer than the SOC follow-up.

To address the high uncertainty in the analysis of base cases, they conducted several analyses relaxing missing data assumption and removing variables that had minimal overlap between the studies. In comparison to the base case, with an ESS of 13.1 and an HR of 0.22, when looking at the missing values in prior lines of therapy and patients’ ECOG performance score, the ESS was 23.6 with an HR of 0.10 (95% CI, 0.04-0.22). Then removing patients with cancer of unknown primary (CUP) tumor and breast cancer, the ESS was 26.4 (HR, 0.10; 95% CI, 0.05-0.21), removing patients with colorectal cancer (CRC) showed the ESS at 37.9 (HR, 0.12; 95% CI, 0.05-0.26), and removing patients with salivary cancer, the ESS ended at 38.6 (HR, 0.12; 95% CI, 0.05-0.25).

A restricted mean survival time analysis was also conducted up to 30.5 months, which was the minimum of the largest observed event time within the SOC arm. The researchers found an estimated 10.8-month mean survival advantage with larotrectinib when compared with SOC (95% CI, 7-14.6; P < .01). On average, this meant that potential patients treated with larotrectinib at 30.5 months of follow-up would survive an additional 10.8 months, compared with SOC. However, the researchers wrote in their poster that the analysis was conservative and that the mean benefit would likely continue to increase with a longer follow-up of patients.

Eighty-two percent of patients were positive for NTRK1 fusion in both arms after adjustments, and 39.3% were either 65 years old or older on the trial. In both arms there was an even split between patients with an ECOG performance score of 0 to 1 and 2 to 4; whereas before matching, 87.1% of patients on larotrectinib had a performance score of 0 to 1.

Several tumor types were observed in the SOC arm, including uterine (4%), biliary (4%), stomach (4%), endometrial (4%), CUP (4%), breast (4%), salivary gland (7%), non–small cell lung cancer (18%), soft tissue sarcoma (21%), and CRC (32%). In the larotrectinib arm, there were no cases of uterine, biliary, stomach, or endometrial cancer. When matching the number of lines of therapy since the patients’ diagnosis, 71.4% of patients had 0-2 lines of therapy and 28.6% had 3 or more lines of therapy since the time of their diagnosis. Most patients (64.3%) had stage III or IV disease at their initial diagnosis, with 17.9% of patients initially diagnosed with stage 0 to II disease. Most patients also had no or unknown brain metastases.

Investigators noted some study limitations, including that MAIC assumes that all effect modifiers are considered, and this analysis was limited to data reported from 2021. However, Bokemeyer concluded there is a clear OS advantage for using larotrectinib in NTRK-positive patients.

Sustained Efficacy Supports Use in CNS Tumors

Larotrectinib (Vitrakvi) continued to produce rapid and durable responses, with a high disease control rate (DCR) and an acceptable toxicity profile in patients with TRK fusion–positive, primary central nervous system (CNS) tumors, according to long-term data from 2 clinical trials presented during the 2022 ASCO Annual Meeting.4

Data from the long-term analysis of those treated on the phase 1/2 SCOUT study (NCT02637687) and the phase 2 NAVIGATE study (NCT02576431) showed that larotrectinib elicited an objective response rate (ORR) of 30% (95% CI, 16%-47%). In 28 pediatric patients, the ORR was 39% (95% CI, 22%-59%). The best responses included complete response (CR) in 8% of patients, partial response (PR) in 22%, stable disease (SD) for 24 weeks or more in 43%, SD for less than 24 weeks in 14%, and progressive disease (PD) in 14%.

A DCR for 24 weeks or more was observed in 73% of patients (95% CI, 56%-86%); in adult patients, this rate was 24% (95% CI, 14%-79%) and in pediatric patients, it was 82% (95% CI, 63%-94%). At a median follow-up of 25.6 months, the median DOR was not reached (95% CI, 4.9-not evaluable [NE]). The 24-month DOR was 53% (95% CI, 23%-83%).

Data from SCOUT and NAVIGATE served as partial basis for an accelerated FDA approval granted in 2018 for larotrectinib as treatment of adult and pediatric patients with solid tumors that have a NTRK gene fusion without a known acquired resistance mutation.

A prior report from these trials showed that larotrectinib achieved an ORR of 30% (95% CI, 16%-49%) and a 24-week DCR of 73% (95% CI, 54%-87%) in 33 evaluable adult and pediatric patients with TRK fusion primary CNS tumors. In addition, treatment with larotrectinib resulted in tumor shrinkage in 82% of patients who had measurable disease.5

The dataset for the long-term analysis included 26 pediatric patients from SCOUT who were under age 21 years with locally advanced metastatic solid tumors or CNS tumors and 12 adult and adolescent patients from NAVIGATE who were 12 years of age or older with advanced solid tumors and TRK fusion cancer. Collectively, the 38 patients were evaluated for the primary end point of investigator-assessed ORR per RECIST 1.1 criteria. The secondary end points assessed for the long-term analysis included duration of response (DOR), progression-free survival (PFS), overall survival (OS), and safety.4

Adult patients were administered oral larotrectinib at 100 mg twice daily, and pediatric patients received 100 mg/m2 twice daily. Patients enrolled had a median age of 10.8 years (range, 1.3-79.0) and 50% were female. The most common type of NTRK gene fusion identified in patients at baseline was NTRK2 in 74%, followed by NTRK1 in 16%, and NTRK3 in 11%. ECOG performance score ranged from 0 to 2; most patients (53%) had a score of 0, 34% had a score of 1, and 11% had a score of 2.

Most patients assessed had high-grade glioma (61%) and low-grade glioma (24%). Sixteen percent of patients had other histologies.

In terms of prior therapies, 71% of patients had undergone surgery, 58% received radiotherapy, and 87% had systemic therapy, and the median number of prior systemic therapies was 1 (range, 0-8). In fact, the majority of patient assessed had a median of 1 prior therapy (42%), whereas 21% had 2, another 21% had 3 or more, and 16% had 0.

On prior therapy, 6% of patient achieved a CR and 3% achieved a PR. The most common response to prior therapy was SD in 36%. Moreover, 30% of patients had PD during prior therapy, and 24% had other responses.

Data from SCOUT and NAVIGATE showed that treatment with larotrectinib lasted for 0.1 months to 38.7 months. The median time to response was 1.9 months (range, 1.0-3.8). Fifty-eight percent of patients progressed on larotrectinib and 4 continued the treatment post-progression for 4 weeks or more.

The median follow-up for PFS assessment was 27.4 months, and larotrectinib achieved a median PFS of 16.5 months (95% CI, 6.7-NE). At 24 months, the PFS rate observed with larotrectinib was 40% (95% CI, 24%-57%). At a median follow-up of 26.7 months, the median OS was not reached (95% CI, 22.6-NE). The 24-months OS rate among the patients evaluated was 65% (95% CI, 47%-83%).

Fifty-five percent of patients in the dataset experienced treatment-related adverse effects (TRAEs), but 86% of them were grade 1 or 2. Eight percent of patients experience grade 3 and 4 TRAEs, which included gamma-glutamytransferase increase, hyperglycemia, hypernatremia, hyponatremia, and neutrophil count decrease. The most common TRAEs of any grade included upper respiratory tract infection, vomiting, diarrhea, alanine aminotransferase increase, headache, anemia, and aspartate aminotransferase increase.

Treatment-related neurologic AEs were also observed in the cohort. Notably, most events were grades 1 and 2 and occurred in the same patients. However, memory impairment was observed in 2 patients, 1 of whom was an adult and other a pediatric patient. No patients in the study discontinued treatment because of a TRAE.4

As data support the use of NTRK inhibitors in practice, diagnostic approaches to identify these fusions should become part of routine practice. Testing for NTRK fusions can be executed using a variety of methods including immunohistochemistry, fluorescence in situ hybridization, reverse transcriptase polymerase chain reaction and next-generation sequencing. Each of these methods provides advantages and disadvantages in terms of cost, array of detectable mutations and alterations, and availability.6


  1. Bokemeyer C, Paracha N, Lassen U, et al. Overall survival (OS) of patients with TRK fusion–positive cancer receiving larotrectinib versus standard of care (SoC): A matching-adjusted indirect comparison (MAIC) using real-world data (RWD). J Clin Oncol. 2022;40(suppl 16):6579. doi:10.1200/JCO.2022.40.16_suppl.6579
  2. Vitrakvi. Prescribing information. Bayer HealthCare Pharmaceuticals Inc; 2021. Accessed June 24, 2022.
  3. Demetri G, Peters S, Davies J, et al. Characteristics and outcomes of patients (pts) with NTRK fusion-positive (NTRK+) metastatic / locally advanced (LA) solid tumours receiving non-TRK inhibitor (TRKi) standard of care (SoC), and prognostic value of NTRK fusions in clinical practice. Ann Oncol. 2021;32(suppl_5):S382-S406. doi:10.1016/ annonc/annonc686
  4. Perreault S, Drilon AE, Lassen LN, et al. Long-term control and safety of larotrectinib in a cohort of adult and pediatric patients with tropomyosin receptor kinase (TRK) fusion primary central nervous system (CNS) tumors. J Clin Oncol. 2022; 40 (suppl 16):2010. doi:10.1200/ JCO.2022.40.16_suppl.2010
  5. Doz F, van Tilburg, Geoerger B, et al. Efficacy and safety of larotrectinib in TRK fusion-positive primary central nervous system tumors. Neuro Oncol. 2022;24(6):997-1007. doi:10.1093/neuonc/ noab274
  6. Penault-Llorca F, Rudzinski ER, Sepulveda AR. Testing algorithm for identification of patients with TRK fusion cancer. J Clin Pathol. 2019;72(7):460-467. doi:10.1136/jclinpath-2018-205679
Related Videos
Shridar Ganesan, MD, PhD
In this fifth episode of OncChats: Assessing the Promise of AI in Oncology, Toufic A. Kachaamy, MD, and Douglas Flora, MD, LSSBB, FACCC, discuss the need for evidence to support the utilization of different artificial intelligence tools in healthcare.
Alan Tan, MD
In this fourth episode of OncChats: Assessing the Promise of AI in Oncology, Toufic A. Kachaamy, MD, and Douglas Flora, MD, LSSBB, FACCC, explain the importance of having a diverse editorial board behind a new journal on artificial intelligence in precision oncology.
Christopher M. Gallagher, MD
Howard S. Hochster, MD
In this third episode of OncChats: Assessing the Promise of AI in Oncology, Toufic A. Kachaamy, MD, and Douglas Flora, MD, LSSBB, FACCC, discuss the launch of a new journal that will focus on educating the field on the latest research efforts being made with artificial intelligence.
In this second episode of OncChats: Assessing the Promise of AI in Oncology, Toufic A. Kachaamy, MD, and Douglas Flora, MD, LSSBB, FACCC, explain how artificial intelligence tools may be leveraged in the oncology field to provide personalized care.
In this first episode of OncChats: Assessing the Promise of AI in Oncology, Toufic A. Kachaamy, MD, and Douglas Flora, MD, LSSBB, FACCC, define artificial intelligence and highlight the goals of these tools in healthcare.
Julia Rotow, MD