Abstract
Imatinib has clinical activity in chronic graft-versus-host disease (cGvHD), a significant complication of allogeneic hematopoietic cell transplant. Nilotinib is a tyrosine kinase inhibitor that targets the same receptors as imatinib but with different affinities. We tested the hypothesis that nilotinib is safe and has clinical activity in cGvHD. 33 participants were enrolled in a phase 1/2 dose escalation and dose extension clinical trial of nilotinib for the treatment of steroid refractory or dependent cGvHD (ClinicalTrials.gov, NCT01155817). We assessed safety, clinical response, and pre-treatment anti-platelet derived growth factor receptor alpha chain (anti-PDGFRA) antibody levels. The 200 mg dose was identified as the maximum tolerated dose and used for the phase 2 dose extension study. At 6 months, the incidence of failure free survival (FFS), cGvHD progression, and nilotinib intolerance resulting in its discontinuation was 50%, 23%, and 23%, respectively. Chronic GvHD responses in skin, joints, and mouth were observed at 3 and 6 months based upon improvement in respective NIH organ severity scores. Pre-treatment anti-PDGFRA antibody levels >0.150 optical density as measured by enzyme linked immunosorbent assay correlated with longer FFS time (P<0.0005) and trended with time until cGvHD progression (P<0.06) but not drug intolerance. Nilotinib may be effective for corticosteroid resistant / refractory cGvHD in some patients but its use is limited by intolerable side effects. Selection of patients with high pre-treatment anti-PDGFRA antibody levels might improve the risk-benefit ratio of nilotinib and better justify its side effects.
Keywords: chronic graft versus host disease, nilotinib, clinical trial
INTRODUCTION
Chronic graft versus host disease (cGvHD) is a major complication of allogeneic hematopoietic cell transplantation (alloHCT). It can result in long term pain and disability and negatively impact quality of life. [1] In vitro and early clinical trial experiences suggest that imatinib may have modest activity in corticosteroid refractory / dependent cGvHD, especially its sclerotic manifestations. [2–9] A recent randomized phase 2 crossover study of imatinib or rituximab for cutaneous sclerotic cGvHD demonstrated a significant clinical response in 26% of the patients initially treated with imatinib. The response lasted longer than 6 months in 17% of the imatinib treated patients. [10]
One possible mechanism of action for imatinib in cGvHD is the inhibition of auto-antibody mediated platelet derived growth factor receptor activation. [11] However, this has not been uniformly corroborated by in vitro studies, and responses to imatinib can occur in patients without anti-platelet derived growth factor receptor alpha chain (anti-PDGFRA) antibodies. [2] These responses suggest that protein receptors targeted by imatinib may be involved in the pathophysiology of cGvHD and that more potent inhibition of these receptors could result in a greater response rate. Nilotinib is a tyrosine kinase inhibitor that targets the same receptors as imatinib (DDR1, DDR2, BCR-ABL, PDGFR alpha/beta, KIT, and NQO2) but with different affinities (half maximal inhibitory concentration (IC50) 3.7, 5.2, 20, 71, 200, and 1800 nM, respectively). [12–14] DDR1, DDR2, and BCR-ABL are more potently inhibited by nilotinib while PDGFR alpha/beta, KIT, and NQO2 are less potently inhibited by nilotinib. We conducted a phase 1/2 clinical trial to test the hypothesis that nilotinib would be active in corticosteroid refractory / dependent cGvHD.
How to determine cGvHD therapeutic responses is greatly debated. Overall impressions of clinicians treating cGvHD patients have been shown to correlate with response. [15, 16] Subsequently, failure free survival (FFS) based upon the initiation of a new systemic therapy has been proposed as an endpoint. [17] Concurrently, the National Institutes of Health (NIH) cGvHD Consensus Committee (NCGCC) has proposed organ specific measurements using predetermined scales followed by summation of measurements from multiple organ systems as another endpoint. [18] A secondary aim of this study was to determine how clinical responses measured with the FFS method relate to those measured with the NIH consensus criteria.
Here we present safety studies and clinical outcomes of nilotinib therapy for corticosteroid refractory / dependent cGvHD using the FFS method and NIH consensus criteria. We also correlate anti-PDGFRA antibodies with clinical outcomes.
MATERIALS AND METHODS
Clinical trial
Eligibility.
All subjects had corticosteroid dependent / refractory cGvHD with manifestations that could be followed on physical examination or with laboratory studies. These manifestations included but were not limited to erythema, hyperpigmentation, lichenification, and ulceration of the skin; moveable or non-moveable sclerosis; fasciitis; oral erythema, hyperkeratosis, ulceration, and mucoceles; bronchiolitis obliterans diagnosed by biopsy or pulmonary function testing with corroborating radiologic findings; and ocular dryness, grittiness, and pain with Schirmer’s testing less than 5 mm after 5 minutes. The organs involved are presented in Table 1. Corticosteroid refractory disease was defined as potentially reversible cGvHD manifestations that persisted despite a prednisone or corticosteroid equivalent dose ≥0.5 mg/kg/d for ≥1 month. Corticosteroid dependence was defined as potentially reversible cGvHD manifestations that persisted despite a prednisone or corticosteroid equivalent dose ≥0.25 mg/kg/d for ≥3 months. Corticosteroids and two other concurrent immunosuppressants were allowed. Previous treatment with imatinib was allowed. There was no limit to the number of cGvHD treatments subjects could receive prior to enrollment.
Table 1.
Subject characteristics (N=33)
| Age in years, median (range) | 50 (22–74) |
| Patient gender, male/female, N | 14/19 |
| Donor type, N | |
| Matched related donor | 16 |
| Matched unrelated donor | 15 |
| Mismatched unrelated donor | 2 |
| Graft source, N | |
| Peripheral blood | 33 |
| Bone marrow | 0 |
| Diagnosis, N | |
| Acute lymphoblastic leukemia | 7 |
| Acute myeloid leukemia or myelodysplastic syndrome | 15 |
| Chronic myeloid leukemia | 1 |
| Chronic lymphocytic leukemia or small cell lymphoma | 3 |
| Hodgkin lymphoma | 1 |
| Myelofibrosis | 1 |
| Multiple myeloma | 1 |
| Non-Hodgkin lymphoma | 3 |
| Prolymphocytic leukemia | 1 |
| Months s/p transplant at enrollment, median (range) | 40.2 (20.6–161.7) |
| Number of systemic cGvHD therapies used prior to nilotinib, median, (range) | 3 (1–7) |
| Systemic cGvHD therapies used prior to nilotinib, N | |
| Prednisone or methylprednisolone | 33 |
| Rituximab | 17 |
| Sirolimus | 13 |
| Extracorporeal photopheresis | 10 |
| Mycophenolate mofetil | 9 |
| Tacrolimus | 9 |
| Cyclosporine | 7 |
| Imatinib | 3 |
| Methotrexate | 2 |
| Hydroxychloroquine | 2 |
| Antithymocyte globulin | 1 |
| Daclizumab | 1 |
| Dasatinib | 1 |
| Pentostatin | 1 |
| Thalidomide | 1 |
| Thoracoabdominal radiation | 1 |
| Chronic GvHD organ involvement upon study enrollment, N | |
| Skin or connective tissue | 32 |
| Ocular | 22 |
| Oropharynx | 21 |
| Lung | 16 |
| Esophagus or lower gastrointestinal tract | 11 |
| Liver (AST, ALT, alkaline phosphatase) | 5 |
Key: cGvHD -chronic graft versus host disease
Design and statistical considerations.
This was a dose escalation trial with 4 sequential dose levels: 200 mg daily, 200 mg twice daily, 400 mg daily, and 400 mg twice daily followed by a dose extension phase at the maximum tolerated dose (MTD). Enrollment followed a standard 3+3 design. [19] Subject demographic and clinical characteristics were summarized with descriptive statistics.
Safety was determined by the observation of adverse events graded according to Common Terminology Criteria for Adverse Events (version 4.0). This trial had the power to detect an adverse event with a true frequency of 10% among 20 patients during the extension phase 88% of the time. A stopping rule terminated accrual to the extension phase if the lower bound of a 1-sided 80% confidence interval for the rate of dose limiting toxicity (DLT) exceeded 20%. This rule was evaluated every 5 patients. Operationally, the stopping rule was triggered if 3 of 5, 4 of 10, 5 of 15, or 6 of 20 patients experienced a DLT. Patients were evaluable for safety after receiving any treatment with nilotinib.
Study participants were treated with nilotinib until their chronic GvHD manifestations progressed and a new systemic immunosuppressant therapy was initiated. Tacrolimus dosing was adjusted to maintain stable serum levels. Because small flares of chronic GvHD can occur which may not represent failures of current therapy and can be managed with temporary increases in corticosteroid dose [20], a temporary increase in steroids was allowed until the symptoms returned to baseline. If criteria were met for progressive cGvHD, nilotinib therapy was categorized as a failure in that subject and the subject was removed from the study. Topical therapy for the eyes, mouth, and skin were allowed at the treating physician’s discretion and not considered to be systemic treatments.
FFS of nilotinib was retrospectively determined based upon the duration of therapy. Subjects discontinued nilotinib (therapeutic failure) for cGvHD progression, drug intolerance, withdrawal of consent by the subject, and malignant relapse. Chronic GvHD progression was defined as worsened disease manifestations requiring new systemic immunosuppressive therapy. Discontinuation of nilotinib due to drug intolerance required documentation of an associated adverse event (AE). Only continued nilotinib administration without clinically significant cGvHD progression represented FFS. FFS was described by the Kaplan Meier estimator. The cumulative incidence for each single cause of nilotinib failure (cGvHD progression, discontinuation of nilotinib due to adverse events or withdrawal of subject consent, and relapse) was calculated using the other causes as competing risks.
Change from baseline in prednisone equivalent dose and total Lee cGvHD symptom scale score at 3, 6, and 12 months of treatment were measured as surrogates for disease response. [21] The Wilcoxon signed-rank test was used to determine the statistical significance of differences between paired before and after treatment measurements of prednisone equivalent dose and the Lee cGvHD scale score.
Chronic GvHD status in specific organ systems was assessed at 1, 2, and 3 months of therapy using 2005 NCGCC scales. [22] These scales were based upon physical exam findings for skin, joint, fascia, and mouth involvement, and based upon provider assessments of symptomology for eye, lung, esophageal, and upper and lower gastrointestinal involvement. Additional restaging was performed on a monthly basis as needed. In 2014, the NCGCC recommended using a subset of the 2005 scales (eliminating those found to be not useful) for which the available data supported their use to determine organ specific and overall responses. [18] The data shown here is a retrospective application of the 2014 criteria to the data elements collected using the 2005 organ specific scales. When taken in sum across all organ systems for a single patient, the NCGCC scales could result in three outcomes: complete response (CR) defined as the resolution of all organ specific manifestations, partial response (PR) defined as improvement in at least one specific organ without worsening in any other, and lack of response defined as no change in any organ or worsening in at least one organ. [18]
Patients were evaluable for response after ≥28 days of therapy. Patients who did not achieve 28 days of therapy were replaced. Corticosteroids were tapered as tolerated while other immunosuppressant dosing was adjusted to maintain therapeutic blood levels. This trial was approved by the respective institutional review boards of the participating institutions and registered at ClinicalTrials.gov as NCT01155817.
Patients.
Thirty-three patients were enrolled from September 7, 2010 to February 11, 2013. Formal follow-up stopped on December 31, 2013 due to the end of funding. Patient characteristics are presented in Table 1. Pre-nilotinib treatment biopsies showed skin thickening consistent with cutaneous cGvHD. Figure 1 shows the course of patients through the trial. Thirty-two subjects received study drug and were evaluable for safety. Thirty subjects received ≥28 days of therapy and were evaluable for response. Median age (range) was 50 (22–74) years. Median time (range) from transplant to study enrollment was 40.2 (20.6–161.7) months and from cGvHD diagnosis was 36.1 (7.5–154) months. The total observation time for all subjects was 264.3 months.
Figure 1. Consort diagram.
Key: * <28 days of treatment, ** Progressive chronic GvHD (9), drug intolerance (14), and recurrent malignancy (2).
Correlative studies.
Serum nilotinib concentrations were determined by LC-MS/MS before and after the daytime dose on days 1 and 8. [23] Anti-PDGFRA antibodies were quantified by enzyme linked immunosorbent assay (ELISA) prior to treatment with nilotinib. [2] Treating physicians did not have access to the anti-PDGFRA antibody levels of subjects while they were enrolled onto the trial.
RESULTS
Toxicity, dose escalation, pharmacokinetics, and relapse
Grade 2–3 adverse events (AEs) possibly or probably related to nilotinib are presented in Table 2. No grade 4–5 AEs attributed to nilotinib were observed. All AEs occurred at the 200 mg daily dose except for two grade 3 AEs which occurred at the 200 mg twice daily dose: lipase elevation, which was asymptomatic, and grade 3 Pneumocystis jiroveci pneumonia. The AEs which resulted in discontinuation of drug were arterial vascular occlusion, lipase elevation, muscle cramping (n=2), myocardial infarction, neutropenia, and suspected fungal pneumonia. Five adverse events were considered serious because they resulted in hospitalization of the patient: aspiration pneumonia, myocardial infarction, pancreatitis, Pneumocystis jiroveci pneumonia, and shingles infection. All were grade 3.
Table 2.
Adverse events possibly and probably related to nilotinib
| Adverse Event | Grade 2* | Grade 3* |
|---|---|---|
| Arterial vascular occlusion | 1 | |
| Aspiration pneumonia | 1‡ | |
| Dehydration | 1 | |
| Hyponatremia | 1 | |
| Hypophosphatem ia | 4 | 3 |
| Influenza B | 1 | |
| Joint stiffness | 1 | |
| Lipase elevation | 1 | 2 |
| Liver function test elevation (ALT, AST, AP, GGT) | 4 | 1 |
| Muscle cramping | 1 | 1 |
| Myocardial infarction | 1‡ | |
| Neutropenia | 1 | |
| Pancreatitis | 1‡ | |
| Pneumocystis jiroveci pneumonia | 1‡ | |
| Prolonged QTc | 1 | |
| Shingles | 1‡ | |
| Suspected fungal pneumonia | 1 | |
| Upper respiratory infection | 1 | |
| Worsening fatigue | 2 |
Number of subjects out of 32 experiencing the adverse event
Serious adverse event resulting in hospitalization
No DLTs were observed during the 200 mg daily cohort and two DLTs occurred at the next dose level of 200 mg twice daily. The AEs which occurred at the 200 mg daily dose did not occur during the dose escalation period or were not severe enough to limit the dose. Therefore, the MTD was defined as 200 mg daily and the maximum administered dose (MAD) was defined as 200 mg twice daily.
The median trough serum nilotinib concentration at the MTD (200 mg once daily) one week after initiation was 769 nM (range 121–2289, n=22). This level is higher than the published in vitro nilotinib IC50s for DDR1, DDR2, BCR-ABL, PDGFR alpha/beta, and KIT (3.7, 5.2, 20, 71, and 200 nM, respectively), but not for NQO2 (1800 nM). [12–14] The median trough serum concentration was higher than previously reported trough concentrations. [24] These trough concentrations suggest the in vivo inhibition of the protein targets of nilotinib except for NQO2. The median trough plasma nilotinib concentration at MAD (200 mg twice daily) one week after initiation was 3426 nM (range 1442–4627, n=3). Both of the DLTs occurred at MAD concentrations of nilotinib.
Two subjects relapsed on days 86 and 359 of nilotinib therapy (days 796 and 1159 after hematopoietic stem cell graft infusion) with T cell prolymphocytic leukemia and Hodgkin lymphoma, respectively.
Responses determined using the FFS method
We retrospectively determined FFS based upon the duration of nilotinib therapy observed for each subject. FFS was 50% (95% confidence interval (CI) 35–72%) at 6 months and 23% (95% CI 12–45%) at 1 year (Figure 2A). Drug intolerance due to AEs or withdrawal of consent was the primary reason for therapeutic failure followed by cGvHD progression and malignant relapse (Table 3). Drug intolerance increased from 23% (95% CI 10–40%) to 40% (95% CI 22–57%) of subjects at 6 months and 2 years, respectively. The cumulative incidence of cGvHD progression plateaued beginning at 6 months and remained stable through 2 years (Figure 2B). This suggests that responses may be durable in the subset of cGvHD remaining on nilotinib at 6 months. Three subjects received imatinib as cGvHD therapy prior to receiving nilotinib. The FFS time for these subjects was 171, 357, and 910 days.
Figure 2. Failure free survival of nilotinib and causes of therapeutic failure.
A. Failure free survival. All subjects (N=30) received > 28 days of therapy. Key: cross – censored, dashed line – upper and lower limits of the 95% confidence interval, solid line – failure free survival. B. Causes of therapeutic failure on nilotinib. All subjects (N=30) received ≥ 28 days of therapy. Key: dashed line – drug intolerance, solid line – chronic graft versus host disease, dotted line – malignant relapse, cross – censored.
Table 3.
Clinical outcomes (n=30)
| Time (Months) | FFS Incidence (95% CI), N | Therapeutic Failure | Censored N | ||
|---|---|---|---|---|---|
| Drug Intolerance Cumulative Incidence (95% CI), N | Chronic GvHD Cumulative Incidence (95% CI), N | Relapse Cumulative Incidence (95% CI), N | |||
| 3 | 77 (63–93), 23 | 13 (4–28), 4 | 7 (1–19), 2 | 3 (0–15) ,1 | 0 |
| 6 | 50 (35–72), 15 | 23 (10–40), 7 | 23 (10–40), 7 | 3 (0–15) ,1 | 0 |
| 12 | 23 (12–45), 6 | 40 (22–57), 12 | 30 (15–47), 9 | 7 (1–20), 2 | 1 |
| 24 | 9 (2–45), 1 | 54 (27–74), 14 | 30 (15–47), 9 | 7 (1–20), 2 | 4 |
Key: CI- confidence interval, FFS- failure free survival (Kaplan Meier estimator), GvHD- graft versus host disease
Surrogates for response
We prospectively assessed the change in prednisone equivalent dose and total Lee cGvHD symptom scale score over time as another indicator for clinical response in those patients who continued to be treated with nilotinib at 3, 6, and 12 months. The median (range) change in prednisone equivalent dose from baseline to 3, 6, and 12 months was −0.01 (−0.15 to 0.33, n=26, p=NS), −0.06 (−0.26 to 0.09, n=14, p<.05), and − 0.08 (−0.32 to 0, n=5, p=NS) mg/kg/day, respectively. Although the difference in prednisone equivalent dose was statistically significant at 6 months, the projected change in clinical dosing based on the median dose for a typical 70 kg person would be −4.2 mg/day. This reduction in dose probably would not represent a clinically significant change.
The median (range) change in Lee cGvHD symptom scale score from baseline to 3, 6, and 12 months was −5 (−24 to 6, n=25, p=0.001 ), −9 (−33 to +1, n=14, p=0.001 ), and −21 (−36 to −9, n=4, p=NS), respectively. Although statistically significant at 3 and 6 months, only the decrease in Lee symptom score at 6 and 12 months would be considered clinically significant (requiring a decrease of at least 6) using established criteria. [21, 25]
Responses based upon the 2014 NCGCC
At 3 months, 26 patients continued on nilotinib while 4 had been removed from the study. Of the 26 continuing patients, 12 (46%) were scored as partial responders while 18 (54%) were scored as non-responders with either no change (n=2) or progression (n=16) of cGvHD organ manifestations (Figure 3A). Responses were seen most frequently in mouth (n=5), skin (n=3), and joint (n=3). Symptomatic improvement occurred in the eye (n=5), and the upper (n=3) and lower (n=3) gastrointestinal tract. At 6 months, 16 patients continued on nilotinib while 14 had been removed from the study. Of the 16 continuing patients, 4 (25%) were scored as partial responders while 12 (75%) were scored as non-responders. All 12 had progression of cGvHD organ manifestations (Figure 3B).
Figure 3. Chronic GvHD organ specific responses.
The response of each organ system to nilotinib is shown as a heat map with each row representing a single subject. Subjects are grouped according to whether they continued treatment with nilotinib (yes or no) at the indicated timepoint and their overall response according to the 2014 NCGCC (PR, NC, or PD). A. Organ specific responses at 3 months. Discrepancies are seen between response classification according to NCGCC and FFS. Heatmaps from the last follow up visit for subjects who are no longer receiving nilotinib at 3 months are also shown. B. Organ specific responses at 6 months. Discrepancies are again seen between response classification according to NCGCC and FFS. Heatmaps from the last follow up visit for subjects who are no longer receiving nilotinib at 6 months are also shown. Key: black – no measurement, blue – complete response, green – partial response, orange – no change, red – progressive disease, cGvHD – chronic graft versus host disease, FFS – failure free survival, GI – gastrointestinal, NC – no change as per NCGCC, NCGCC – NIH chronic graft versus host disease consensus criteria (2014), NIH – National Institutes of Health, no – subject does not continue to receive nilotinib, PD – progressive cGvHD as per NCGCC, PR – partial response as per NCGCC, yes – subject continues to receive nilotinib.
Clinical outcomes and anti-PDGFRA antibodies
ELISA based optical density (OD) measurements of pre-treatment blood anti-PDGFRA antibody concentration ranged from −0.024 to 1.754 with a median of 0.1475. We rounded this value to 0.150 for convenience in the subsequent analyses. This chosen value did not change the rank order of anti-PDGFRA antibody concentration measurements: 50% of patients tested for pre-treatment anti-PDGFRA antibodies had levels ≥0.150 OD. FFS was significantly associated with pre-treatment anti-PDGFRA antibody levels ≥0.150 OD in subjects receiving ≥28 days of therapy (n=30, P<0.0005, Figure 4A). Treating physicians did not have knowledge of the anti-PDGFRA antibody levels. When causes of nilotinib failure were analyzed, only cGvHD progression trended towards being inversely correlated with anti-PDGFRA antibodies ≥ 0.150 OD (P<0.06, Figure 4B-C, data for relapse not shown due to small (n=2) sample size).
Figure 4. Correlation between anti-PDGFRA antibodies and nilotinib clinical outcomes.
A. Pre-treatment anti-PDGFRA antibody levels correlate with failure free survival. B. Pre-treatment anti-PDGFRA antibody levels trend towards correlation with time to cGvHD progression. C. Pre-treatment anti-PDGFRA antibody levels do not correlate with drug intolerance. Key: solid line – anti-PDGFRA antibodies ≥0.150 OD, dashed line – anti-PDGFRA antibodies ≥ 0.150 OD, cross – censored observations, anti-PDGFRA – anti-platelet derived growth factor receptor alpha chain, OD – optical density as measured by enzyme linked immunosorbent assay.
DISCUSSION
We conducted a multicenter unblinded phase 1 / 2 dose escalation and extension trial testing the hypothesis that nilotinib is clinically effective therapy for corticosteroid refractory / dependent cGvHD. The 200 mg daily and twice daily doses of nilotinib were tested. The 200 mg daily dose was safe to administer after allogeneic hematopoietic cell transplant albeit with high rates of intolerance. Based upon response measured using a variety of measurement tools (FFS, NIH consensus criteria, and Lee cGvHD symptom scale score), we conclude that nilotinib has clinical activity in chronic GvHD. This response was consistent with those previously seen with imatinib, which targets the same cellular receptors but with different affinities. [2] The median FFS time was longer for subjects previously treated with imatinib than for all study participants, suggesting that nilotinib may have activity in cases of imatinib failure.
We observed a 6 month FFS of 50% which was comparable to the 6 month FFS of 56% reported from a single center retrospective analysis of 312 patients receiving unspecified second line systemic cGvHD therapies. [17] The slight difference in FFS may have been due to a more heavily pre-treated population in this study.
We observed that continued FFS did not correlate with disease response according to the NCGCC. This may have been due to the increased granularity of the NCGCC which prespecifies the changes that would constitute response or progression as compared to FFS for which response or progression are dependent on the clinician’s overall assessment of the patient. Although many of the individual NCGCC organ specific scales are considered by expert opinion to be clinically meaningful, the magnitude of these changes, particularly when considered across several involved organs, may not have been sufficient to change a patient’s treatment plan and result in a change in FFS. [26–29] The discrepancy between responses classified by FFS and NCGCC could also reflect a placebo effect from patient and physician enthusiasm for the perceived efficacy of the study drug, resulting in a greater willingness to tolerate short term worsening in disease status for possible future improvements. This possible bias could have translated into a delay in the initiation of a tertiary systemic therapy and a longer FFS.
Changes in the Lee cGVHD symptom scale score were consistent with successful treatment with nilotinib over time and the absence of progressive cGvHD. However, the placebo effect may have affected patient responses captured by the survey instrument. In addition, because of patient drop out due to progressive cGVHD symptoms or intolerance, later time points would be enriched for patients responding to nilotinib further biasing Lee cGvHD scale scores.
Intolerance of nilotinib was a significant issue in this study. We observed that toxicity and nilotinib serum levels correlated with the daily dose. Serum concentrations of nilotinib were higher than expected based upon previous reports. [24] These higher concentrations may have been due to the common post-alloHCT concomitant use of voriconazole or fluconazole which inhibit CYP3A4 reducing the metabolism of nilotinib. Alternatively, the higher concentrations could be due to higher concentrations of alpha-1-acid glycoprotein, an acute phase plasma protein known to strongly bind nilotinib, which may be induced in inflammatory conditions such as GvHD, and which would increase the total serum concentration of nilotinib. Higher than expected serum levels may have increased the number of subjects discontinuing therapy due to adverse events and intolerance.
Pre-treatment anti-PDGFRA antibody levels correlated with FFS duration. This observation differed from previous reports that anti-PDGFRA antibodies did not correlate with response to imatinib and may have been due to more granular data which enabled the detection of the association between anti-PDGFRA antibodies and FFS. [2] The current study examined 30 versus 11 patients in the imatinib study and measured response to nilotinib with greater precision using time to nilotinib failure rather than response at 24 weeks and last follow-up. High pre-treatment anti-PDGFRA antibody levels could be clinically relevant for identifying a subpopulation of cGvHD patients that is more likely to respond to nilotinib, better justifying the risk of adverse side effects. One potential criticism of this analysis is the strong influence of the treating physician on the duration of FFS. However, anti-PDGFRA antibody levels were obtained before nilotinib treatment and investigators did not know the level while subjects were treated with nilotinib, making it unlikely that the anti-PDGFRA level affected the assessment of response to nilotinib.
Despite a statistically significant association with FFS, anti-PDGFRA antibodies only showed a trend towards association with cGvHD progression when specific causes of FFS were examined. Atypical manifestations of cGvHD progression like myalgia and cramping may have been misattributed to drug intolerance rather than cGvHD progression. This misattribution may have altered the strength of the correlation between anti-PDGFRA antibody levels and time to cGvHD progression. It may have also falsely elevated the incidence of adverse events.
In summary, a 200 mg daily dose of nilotinib has an acceptable safety profile and may have clinical activity in steroid refractory / dependent cGvHD. However, its clinical use is limited by intolerance of the drug. The strengths of this trial were demonstration of response with a variety of measurement tools, the use of standardized instruments whose data could be reassessed using new definitions of response as they evolve such as the change from the 2005 to 2014 NCGCC, and the acquisition of pharmacokinetic and toxicity data for nilotinib. The weakness of this trial was the uncontrolled design. Future trials of nilotinib could utilize a blinded study design to control for any placebo effects. Additionally, future trials could utilize a nilotinib dose ≤ 200 mg to improve tolerability. The IC50 of the target receptors were much lower than the observed serum nilotinib levels at the 200 mg dose, suggesting that clinical efficacy could be maintained. Finally, selecting for patients with sclerotic chronic GVHD or pre-treatment anti-PDGFRA antibody levels ≥0.150 may result in a greater response rate, justifying the risk of intolerance to nilotinib. It is not known if the possible predictive value of anti-PDGFRA antibodies would extend to imatinib or other cGvHD therapies. Further testing and validation in a large prospective cohort of patients is needed.
HIGHLIGHTS.
Nilotinib has clinical activity in steroid refractory / dependent cGvHD.
Anti-PDGFRA antibodies may predict which patients respond to nilotinib.
ACKNOWLEDGEMENTS
Novartis supported the clinical trial and correlative studies. This project used the University of Pittsburgh Cancer Pharmacokinetics and Pharmacodynamics Facility supported by P30CA047904 and Roswell Park Core Facilities supported by P30CA016056.
We wish to thank all the patients, BMT nurses, patient coordinators, and staff at all participating centers who made this work possible.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES
- [1].Pidala J, Kurland B, Chai X, Majhail N, Weisdorf DJ, Pavletic S, et al. Patient-reported quality of life is associated with severity of chronic graft-versus-host disease as measured by NIH criteria: report on baseline data from the Chronic GVHD Consortium. Blood. 2011;117:4651–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Chen GL, Arai S, Flowers ME, Otani JM, Qiu J, Cheng EC, et al. A phase 1 study of imatinib for corticosteroid-dependent/refractory chronic graft-versus-host disease: response does not correlate with anti-PDGFRA antibodies. Blood. 2011;118:4070–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Magro L, Mohty M, Catteau B, Coiteux V, Chevallier P, Terriou L, et al. Imatinib mesylate as salvage therapy for refractory sclerotic chronic graft-versus-host disease. Blood. 2009;114:719–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Majhail NS, Schiffer CA, Weisdorf DJ. Improvement of pulmonary function with imatinib mesylate in bronchiolitis obliterans following allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2006;12:789–91. [DOI] [PubMed] [Google Scholar]
- [5].Stadler M, Ahlborn R, Kamal H, Diedrich H, Buchholz S, Eder M, et al. Limited efficacy of imatinib in severe pulmonary chronic graft-versus-host disease. Blood. 2009;114:3718–9. [DOI] [PubMed] [Google Scholar]
- [6].de Masson A, Bouaziz JD, Peffault de Latour R, Wittnebel S, Ribaud P, Rubio MT, et al. Limited efficacy and tolerance of imatinib mesylate in steroid-refractory sclerodermatous chronic GVHD. Blood. 2012;120:5089–90. [DOI] [PubMed] [Google Scholar]
- [7].Baird K, Comis LE, Joe GO, Steinberg SM, Hakim FT, Rose JJ, et al. Imatinib Mesylate for the Treatment of Steroid-Refractory Sclerotic-Type Cutaneous Chronic Graft-versus-Host Disease. Biol Blood Marrow Transplant. 2015;21:1083–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Olivieri A, Cimminiello M, Corradini P, Mordini N, Fedele R, Selleri C, et al. Long-term outcome and prospective validation of NIH response criteria in 39 patients receiving imatinib for steroid-refractory chronic GVHD. Blood. 2013;122:4111–8. [DOI] [PubMed] [Google Scholar]
- [9].Olivieri A, Locatelli F, Zecca M, Sanna A, Cimminiello M, Raimondi R, et al. Imatinib for refractory chronic graft-versus-host disease with fibrotic features. Blood. 2009;114:709–18. [DOI] [PubMed] [Google Scholar]
- [10].Arai S, Pidala J, Pusic I, Chai X, Jaglowski S, Khera N, et al. A Randomized Phase II Crossover Study of Imatinib or Rituximab for Cutaneous Sclerosis after Hematopoietic Cell Transplantation. Clinical Cancer Research. 2016;22:319–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Svegliati S, Olivieri A, Campelli N, Luchetti M, Poloni A, Trappolini S, et al. Stimulatory autoantibodies to PDGF receptor in patients with extensive chronic graft- versus-host disease. Blood. 2007;110:237–41. [DOI] [PubMed] [Google Scholar]
- [12].Day E, Waters B, Spiegel K, Alnadaf T, Manley PW, Buchdunger E, et al. Inhibition of collagen-induced discoidin domain receptor 1 and 2 activation by imatinib, nilotinib and dasatinib. European journal of pharmacology. 2008;599:44–53. [DOI] [PubMed] [Google Scholar]
- [13].Weisberg E, Manley PW, Breitenstein W, Bruggen J, Cowan-Jacob SW, Ray A, et al. Characterization of AMN107, a selective inhibitor of native and mutant Bcr-Abl. Cancer Cell. 2005;7:129–41. [DOI] [PubMed] [Google Scholar]
- [14].Rix U, Hantschel O, Durnberger G, Remsing Rix LL, Planyavsky M, Fernbach NV, et al. Chemical proteomic profiles of the BCR-ABL inhibitors imatinib, nilotinib, and dasatinib reveal novel kinase and nonkinase targets. Blood. 2007;110:4055–63. [DOI] [PubMed] [Google Scholar]
- [15].Curtis LM, Grkovic L, Mitchell SA, Steinberg SM, Cowen EW, Datiles MB, et al. NIH response criteria measures are associated with important parameters of disease severity in patients with chronic GVHD. Bone marrow transplantation. 2014;49:1513–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Arai S, Jagasia M, Storer B, Chai X, Pidala J, Cutler C, et al. Global and organ-specific chronic graft-versus-host disease severity according to the 2005 NIH Consensus Criteria. Blood. 2011;118:4242–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Inamoto Y, Storer BE, Lee SJ, Carpenter PA, Sandmaier BM, Flowers ME, et al. Failure-free survival after second-line systemic treatment of chronic graft-versus-host disease. Blood. 2013;121:2340–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Lee SJ, Wolff D, Kitko C, Koreth J, Inamoto Y, Jagasia M, et al. Measuring therapeutic response in chronic graft-versus-host disease. National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: IV. The 2014 Response Criteria Working Group report. Biol Blood Marrow Transplant. 2015;21:984–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Storer BE. Design and analysis of phase I clinical trials. Biometrics. 1989;45:925–37. [PubMed] [Google Scholar]
- [20].Martin PJ, Lee SJ, Przepiorka D, Horowitz MM, Koreth J, Vogelsang GB, et al. National Institutes of Health Consensus Development Project on Criteria for Clinical Trials in Chronic Graft-versus-Host Disease: VI. The 2014 Clinical Trial Design Working Group Report. Biol Blood Marrow Transplant. 2015;21:1343–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Lee S, Cook EF, Soiffer R, Antin JH. Development and validation of a scale to measure symptoms of chronic graft-versus-host disease. Biol Blood Marrow Transplant. 2002;8:444–52. [DOI] [PubMed] [Google Scholar]
- [22].Pavletic SZ, Martin P, Lee SJ, Mitchell S, Jacobsohn D, Cowen EW, et al. Measuring therapeutic response in chronic graft-versus-host disease: National Institutes of Health Consensus Development Project on Criteria for Clinical Trials in Chronic Graft-versus-Host Disease: IV. Response Criteria Working Group report. Biol Blood Marrow Transplant. 2006;12:252–66. [DOI] [PubMed] [Google Scholar]
- [23].Parise rA Egorin MJ, Christner SM Shah DD, Zhou W Beumer JH. A high-performance liquid chromatography-mass spectrometry assay for quantitation of the tyrosine kinase inhibitor nilotinib in human plasma and serum. J Chromatogr B Analyt Technol Biomed Life Sci. 2009;877:1894–900. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Tanaka C, Yin OQ, Sethuraman V, Smith T, Wang X, Grouss K, et al. Clinical pharmacokinetics of the BCR-ABL tyrosine kinase inhibitor nilotinib. Clinical pharmacology and therapeutics. 2010;87:197–203. [DOI] [PubMed] [Google Scholar]
- [25].Sa Mitchell, Leidy NK Mooney KH, Dudley WN Beck SL, LaStayo PC, et al. Determinants of functional performance in long-term survivors of allogeneic hematopoietic stem cell transplantation with chronic graft-versus-host disease (cGVHD). Bone Marrow Transplant. 2010;45:762–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Inamoto Y, Pidala J, Chai X, Kurland BF, Weisdorf D, Flowers ME, et al. Assessment of joint and fascia manifestations in chronic graft-versus-host disease. Arthritis & rheumatology. 2014;66:1044–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Jacobsohn DA, Kurland BF, Pidala J, Inamoto Y, Chai X, Palmer JM, et al. Correlation between NIH composite skin score, patient-reported skin score, and outcome: results from the Chronic GVHD Consortium. Blood. 2012;120:2545–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Elad S, Zeevi I, Or R, Resnick IB, Dray L, Shapira MY. Validation of the National Institutes of Health (NIH) scale for oral chronic graft-versus-host disease (cGVHD). Biol Blood Marrow Transplant. 2010;16:62–9. [DOI] [PubMed] [Google Scholar]
- [29].Inamoto Y, Chai X, Kurland BF, Cutler C, Flowers ME, Palmer JM, et al. Validation of measurement scales in ocular graft-versus-host disease. Ophthalmology. 2012;119:487–93. [DOI] [PMC free article] [PubMed] [Google Scholar]