CHICAGO—“Inflammatory myofibroblastic tumors (IMFTs) are rare mesenchymal neoplasms that can cause significant functional disabilities, organ dysfunction, and death,” stated Patrick Schöffski, MD, MPH, Head of the Department of General Medical Oncology at the Leuven Cancer Institute, University Hospitals Leuven, Belgium, and principal investigator in the Laboratory of Experimental Oncology at the Catholic University of Leuven (KU Leuven). “These tumors, which occur in children and adult patients, are commonly found in the thorax, retroperitoneal, or pelvic regions. IMFTs show limited sensitivity to conventional chemotherapy and radiotherapy, and there is no drug approved specifically for this indication by regulatory agencies, thus surgical resection is the main method of treating these patients.
“Surgical treatment, which often entails amputations and other debilitating interventions, is the mainstay of treatment; however, complete resection is often not possible because of the proximity of the tumors to vital organs,” he noted. “Unfortunately, IMFTs are characterized by low surgical cure rates, a high rate of histological misclassification, and high rates of local recurrence and invasion, although distant metastasis is less common. Thus, patients with unresectable or locally recurring disease have few treatment options.”
Approximately half of all IMFTs have translocations of the ALK gene. Additionally, some research has shown that another subset of IMFTs have ALK fusions with non-canonical breakpoints, which implies that aberrant ALK signaling may be occurring in a majority of these neoplasms.
“The NCCN began recommending the off-label use of the tyrosine kinase inhibitor crizotinib as a treatment for ALK-positive IMFT a few years ago, after a single ALK-positive patient was reported to benefit,” said Schöffski. Crizotinib, which is an ATP-competitive tyrosine kinase inhibitor (TKI) of ALK and ROS1 (c-ros oncogene 1), has received FDA approval for the treatment of patients having ALK-positive non-small cell lung cancer (NSCLC).
To evaluate the use of crizotinib in adults with IMFTs, the two-stage phase II EORTC (European Organization for Research and Treatment of Cancer) 90101 CREATE trial was initiated. This study, which was biomarker-driven, single-drug, non-randomized, and open-label, recruited patients from five university hospitals and eight specialty clinics in eight European countries, including Belgium, France, Germany, Italy, Netherlands, Poland, Slovakia, and the U.K. At the recent 2018 AACR Annual Meeting, Schöffski presented results for this clinical trial. Concurrent to that presentation was the publication of trial results in Lancet Respiratory Medicine (2018; http://dx.doi.org/10.1016/S2213-2600(18)30116-4).
Concerning the results of this study, Schöffski noted, “Our data in predominantly adult patients with IMFTs, combined with recently published data for children with this disease, support the recommendation to consider crizotinib as the standard of care for patients with ALK-positive IMFT who cannot be treated with surgery.”
Patient enrollment was accomplished using a multi-step registration process. In the first step, prerequisites for registration included a local diagnosis of advanced or metastatic IMFT deemed incurable by surgery, radiotherapy, or systemic therapy; the availability of a formalin-fixed paraffin-embedded tissue sample from either the primary or metastatic tumor site for trial purposes; and written informed consent from the patient for central tissue collection and all other trial-related procedures. Criteria for step 2 included receipt of the tissue with presence of tumor by a central biorepository and confirmation of the correct IMFT diagnosis of by central reference pathology. The reference pathology review was performed using pre-specified morphological and immunohistochemical (IHC) diagnostic criteria for IMFT.
If patients met the criteria for the previous steps, treatment was initiated if the following additional criteria were met: ages ≥ 15 years; an ECOG performance status of 0-2; adequate hematological, renal, and liver function; measurable disease using RECIST v1.1; no previous exposure to crizotinib or other ALK-based inhibitors (all other local or systemic therapies for IMFT were allowed); no acute or chronic severe gastrointestinal conditions; and no current congestive heart failure or cardiac dysrhythmias of grade 2 or more (as per NCI Common Toxicity Criteria), including uncontrolled atrial fibrillation.
The documented presence of specific ALK alteration was not necessary for patient entry into step 3, the clinical screening and treatment phase. Central molecular or genetic analysis was performed while patients were already receiving therapy. Patients were deemed to have ALK-positive tumors if at least 15 percent of their tumor cells showed gene rearrangement using fluorescence in-situ hybridization (FISH) or IHC or both. IHC and molecular analyses were performed centrally at the University Hospitals Leuven, using on average four unstained 4 μm slides from archived primary or metastatic formalin-fixed paraffin-embedded tissue samples provided by the local clinicians to the central biorepository in Milan.
Eligible patients having centrally confirmed IMFT diagnoses were treated with oral crizotinib capsules at a starting dose of 250 mg twice per day (the capsules contained 200 mg or 250 mg of study medication).
“The dosage was based on the approved crizotinib adult dose for ALK-positive NSCLC and there were no dose increases in the protocol; crizotinib was administered at approximately the same times each day on a continuous daily dosing schedule (i.e., no break in dosing), in the absence of drug-related toxicity,” Schöffski explained. If necessary, dose reductions could be performed using 200 mg crizotinib capsules. Treatment was continued until documented disease progression, unacceptable toxicity, or patient refusal was noted.
The primary endpoint for this trial was the proportion of patients achieving an objective response (i.e., either a partial (PR) or complete response (CR)) at any time during treatment as per RECIST 1.1 with response confirmation, as assessed by the local investigator or radiologist.
“This endpoint was chosen based on the response pattern with crizotinib in the labelled indication of NSCLC and because there were no reliable reference data on progression-free survival (PFS) or PFS rate in this rare disease when the protocol was written,” Schöffski explained. Secondary endpoints included duration of response, the proportion of patients achieving disease control (patients displaying stable disease (SD), PR, or CR), PFS, PFS rate, overall survival, and safety.
From October 2012 to April 2017, 35 patients with locally diagnosed IMFT were recruited for this study. Of these, 24 were found to have centrally confirmed IMFT. Among the misclassified cases were one case of myoepithelioma, one presumed calcifying fibrous pseudotumor, and seven not matching the pre-specified morphological and IHC criteria.
In June 2017, without having reached the recruitment maximum of 35 ALK-positive patients, recruitment to both subcohorts of the trial was suspended because patient entry was slow due to the low incidence of IMFT, and because both stage I and stage II success criteria of the trial had already been met.
Of the 24 patients with centrally confirmed diagnoses of IMFT, 16 (67%) were defined as ALK-positive using the protocol criteria (11- FISH and IHC confirmation, 3-IHC only confirmation, 2-FISH only confirmation). Eight patients (33%) were ALK-negative. From this pool of patients with confirmed IMFT diagnoses, 12 with ALK-positive disease and eight with ALK-negative disease entered the treatment phase.
As of the data cutoff date of Nov. 9, 2017, the median follow-up time was 863 days (IQR 358-1304), and of the 20 treated patients, seven (35%) were still being actively treated with crizotinib. A total of 13 patients discontinued crizotinib therapy (65%).
“Six of 12 ALK-positive patients (50%, 95% CI: 21.1-78.9%) and one of seven ALK-negative patients (14%, 0.0-57.9%) achieved an objective response; the median duration of response was 9.0 months in the ALK-positive patients (range 1.4-41.6, IQR 4.4-37.3 months),” Schöffski noted.
“In the lone responding ALK-negative patient, the duration of response was 7.6 months. Notably, all ALK-positive patients had either a confirmed complete or partial response or stable disease according to RECIST as best response, and on Nov. 9, 2017, four ALK-positive responders were still responding and receiving active treatment.”
“Our study highlights how identifying the genetic drivers of a rare type of cancer can be used to find a new precision medicine for patients who otherwise have few treatment options,” Schöffski commented. “The inclusion in our trial of a group of patients with ALK-negative inflammatory myofibroblastic tumors provides valuable insight into the selectivity of crizotinib and our understanding of this rare disease, even if we cannot formally compare the outcomes for the ALK-positive and ALK-negative groups.
“One partial response and three patients with objective shrinkage of their target lesions were noted in tumors defined as ALK-negative according to criteria in this trial,” he stated, further adding, “ROS1 alterations were absent and could not explain these results.
“Perhaps the defined cutoff of 15 percent ALK-positive cells by FISH and IHC may have led to false-negative results in tumors with borderline percentages of positive cells (e.g., 10-20%),” Schöffski noted. “FISH may miss complex rearrangements generating an atypical signal; additionally, the antibody used, clone CD246, DAKO, does not detect all fusion forms of ALK, e.g., EML4-ALK. Atypical breakpoints in ALK-related fusions are not covered by used FISH assay, e.g., FN1-ALK, and additionally, ALK activation may occur due to point mutations instead of rearrangement, as is seen in NSCLC. Further complicating the issue is the fact that IMFTs are known to have other crizotinib-sensitive alterations, e.g., ROS1 rearrangement.”
When asked about the limitations of this trial, Schöffski noted, “Because IMFT is an orphan malignancy, it took 61 months and 13 sites in eight European countries to find 35 patients with a local diagnosis of IMFT and, as a result, this trial had a small sample size and early discontinuation of recruitment. It may not be possible to conduct larger randomized trials; this was the largest prospective treatment series of IMFT cases in the literature and likely provides the best type of evidence that can be generated in this indication.”
Regarding the high number of misdiagnoses by local investigators, Schöffski stated “Considering the observed local misclassification rate of 31.4 percent, reference pathology will be essential for all future trials in this indication.”
When asked to discuss the findings in this trial, Schöffski noted, “66.7 percent of advanced, inoperable IMFTs are ALK-positive according to the criteria used in this trial and concordance between FISH and IHC is seen in 68.7 percent of ALK-positive cases. Crizotinib has important, clinically relevant activity in advanced, inoperable, metastatic IMFT, as the drug achieved 50 percent objective responses (PR, CR) and long-lasting disease control in ALK-positive patients. All predefined success criteria of EORTC 90101 “CREATE” were met, with all ALK-positive participants attaining disease control (CR, PR, or SD) with crizotinib; 73.3 percent are progression-free at 1 year and 81.8 percent are alive 2 years after starting treatment.
“Sporadic activity can also be observed in ALK-negative cases, highlighting the potential relevance of other ALK alterations, alternative molecular drivers affected by the TKI or limitations of assays and criteria used for defining ALK positivity in this trial.”
Regarding the investigational treatment, he noted, “Crizotinib was generally well-tolerated in IMFT patients with high-dose intensity over a long period of time; additionally, the quality and quantity of responses in ALK-positive IMFT participants was consistent with the activity seen in ALK-positive NSCLC patients.” Continuing, he stated, “Our data were in line with recent findings from the pediatric dose finding study in children with ALK-positive IMFT.”
When asked if acquired resistance to crizotinib therapy was observed in this trial, Schöffski replied, “Some patients had secondary RECIST disease progression after achieving disease control (CR, PR, or SD) with crizotinib. There must have been some secondary, acquired resistance; however, we did not investigate the according mechanisms for that yet.”
Regarding the nature of future clinical studies for IMFT, Schöffski speculated, “Molecular studies could explore the sensitivity and resistance mechanisms related to the use of tyrosine kinase inhibitors such as crizotinib. In addition, the definition of ALK-negative disease and role of crizotinib in ALK-negative patients remains to be further elucidated.
“In the absence of validated treatment alternatives for IMFT, and considering the limitations of non-randomized drug testing, we propose to define crizotinib as the standard of care for patients with locally advanced or metastatic ALK-positive IMFT.”
Richard Simoneaux is a contributing writer.