Original Article

Preliminary Characterization of Oral Lesions Associated With Inhibitors of Mammalian Target of Rapamycin in Cancer Patients Stephen Sonis, DMD, DMSc1,2; Nathaniel Treister, DMD, DMSc1,2; Sant Chawla, MD3; George Demetri, MD4,5; and Frank Haluska, MD, PhD6

BACKGROUND: Mammalian target of rapamycin (mTOR) inhibitors may have efficacy as an intervention for advanced malignancies. Oral ulceration (OU), reported as mucositis, has been a dose-limiting toxicity for this new class of agents. An analysis of the appearance, course, and toxicity associations of mTOR inhibitor-associated stomatitis (mIAS) demonstrated that the condition is distinct from conventional mucositis (CM) and more closely resembles aphthous stomatitis. METHODS: Safety data from 78 solid tumor patients enrolled in 2 Phase 1, multicenter trials of the mTOR inhibitor deforolimus (AP23573, MK-8669) were evaluated. Adverse events (AEs) based on National Cancer Institute Common Toxicity Criteria for National Cancer Institute Common Terminology Criteria for Adverse Events (version 3.0) criteria were coded, consolidated, and stratified according to the presence or absence and duration of concordant OU. The relation between OU and other AEs was analyzed. RESULTS: Treatment-emergent AEs were reported in 91% of 78 study participants. OUs were reported in 66%, appeared within 5 days of deforolimus administration, and were discrete, ovoid, superficial, well demarcated, and surrounded by an erythematous halo. Their clinical appearance and distribution were similar to that of aphthous stomatitis but inconsistent with CM. Patients with OU were more likely to have nonspecific rashes and acneiform dermatitis but not gastrointestinal AEs. CONCLUSIONS: OU associated with mTOR inhibitor therapy differed from CM. Lesions more closely resembled those of aphthous stomatitis. The lack of other gastrointestinal involvement but the presence of a higher incidence of concomitant cutaneous AEs provided additional evidence to suggest a distinction between mIAS and CM. Treatment strategies for aphthous stomatitis may be a rational approach for the prevention and control of mIAS. Cancer 2010;116:210–5. C 2010 American Cancer Society. V KEYWORDS: mammalian target of rapamycin inhibitor, mucositis, aphthous stomatitis, cancer.

Because of its role as a central regulator of key pathways believed to be important in the proliferation of cancers, the mammalian target of rapamycin (mTOR) has emerged as a viable target for the generation of a new group of antitumor agents. mTOR inhibitors have demonstrated encouraging efficacy results in patients with a range of advanced malignancies.1 One of these agents, temsirolimus, is approved for the management of renal cell carcinoma.2 Although generally well tolerated, the most common dose-limiting toxicity (DLT) for mTOR inhibitors has been mouth ulcers.3,4 Because mouth ulcers in cancer trials are defined by the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) (version 3.0) as mucositis, it largely was assumed that mTOR-associated oral lesions were equivalent to classic chemotherapy-induced oral mucositis in their clinical presentation, etiopathology, risk, course, and susceptibility for intervention.

Corresponding author: Stephen Sonis, DMD, DMSc, Division of Oral Medicine and Dentistry, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115; Fax: (617) 525-6899; [email protected] 1 Division of Oral Medicine and Dentistry, Brigham and Women’s Hospital and Dana-Farber Cancer Institute, Boston, Massachusetts; 2Department of Oral Medicine, Infection, and Immunity, Harvard School of Dental Medicine, Boston, Massachusetts; 3Sarcoma Oncology Center, Santa Monica, California; 4Center for Sarcoma and Bone Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts; 5Department of Medicine, Harvard Medical School, Boston, Massachusetts; 6Ariad Pharmaceuticals Inc, Cambridge, Massachusetts

We thank Drs. Ratain and Desai (University of Chicago, Chicago, Ill) who served as the principal investigators for the AP23573-02-01 trial, Drs. Tochler and Rowinsky (Cancer Therapy and Research Center, San Antonio, Tex) who served as the principal investigators for the AP23573-02-102 trial, and the numerous coinvestigators and subinvestigators who participated in these studies. DOI: 10.1002/cncr.24696, Received: February 28, 2009; Revised: March 23, 2009; Accepted: April 9, 2009, Published online October 27, 2009 in Wiley InterScience (www.interscience.wiley.com)

210

Cancer

January 1, 2010

mTOR Inhibitor-Associated Stomatitis/Sonis et al

The clinical course of oral mucositis associated with conventional, cycled cancer chemotherapy regimens has been well described.5 Depending on the anticancer drugs used, approximately 10% to 30% of patients who are receiving treatment for solid tumors reportedly develop ulcerative mucositis (NCI-CTCAE grade 2 or worse) during the first cycle of chemotherapy. Among patients for whom there is no dose de-escalation, the incidence of mucositis increases significantly in subsequent treatment cycles. The incidence is as high as 70% in patients who undergo myeloablative conditioning before hematopoietic stem cell transplantation.6 Mucositis typically begins within 4 to 5 days after chemotherapy, peaks between Day 7 and Day 10, and spontaneously resolves soon thereafter.5 Lesions appear on the movable mucosa and are characterized by their nonuniform shape, depth, the presence of a fibrinous pseudomembrane with cellular debris, and lack of peripheral erythema. The ulcers can produce pain of such intensity as to require changes in diet and systemic analgesic intervention. Other than palliation, to the best of our knowledge, there currently is no approved intervention for the prevention or treatment of chemotherapy-induced mucositis in patients with solid tumors. The solid organ transplantation literature often cites ‘‘aphthous-like’’ oral lesions associated with sirolimus (Rapamycin) therapy, an mTOR inhibitor used for its immunomodulatory properties.7 Unlike typical chemotherapy-induced mucositis, aphthous lesions present as discrete, ovoid, relatively shallow ulcers surrounded by a characteristic erythematous margin.8,9 It is interesting to note that their pathogenesis has been studied reasonably well, although it is not entirely understood and differs considerably from that of mucositis.8,10 From a clinical standpoint, a range of effective medical treatments, including topical and systemic steroids, topical leukotriene inhibitors, and inhibitors of tumor necrosis factor, are available.8,11 The objectives of the current study were to define more accurately and to characterize the nature of mTOR inhibitor-associated oral toxicities. We evaluated the clinical features and course of oral lesions and their association with other toxicities among cancer patients who were receiving mTOR inhibitor therapy, and we compared those findings with the expected results among individuals who received conventional chemotherapy. Our findings suggest that mTOR inhibitor-associated stomatitis (mIAS) is distinct from conventional chemotherapyinduced mucositis. It is worth noting that this finding indicates that there may be effective approaches to manag-

Cancer

January 1, 2010

ing this dose-limiting toxicity, thus allowing maximum benefit from this emerging class of anticancer agents.

MATERIALS AND METHODS Interim safety data were obtained from 2 open-label, phase 1, multicenter trials12,13 of deforolimus (AP23573) for patients with solid malignancies (Table 1) who had experienced treatment failure with prior standard therapy or for whom adequate therapy was not available (National Clinical Trials (NCT) identifiers: NCT00060632 for Trial AP23573-02-101 and NCT00060645 for Trial AP23573-02-102). Patients with primary central nervous system malignancy or any form of leukemia were excluded. Patients were excluded from the study if they had received any prior therapy within 4 weeks of deforolimus dosing (6 weeks for mitomycin C and nitrosureas) with the exception of any agents with a known half-life 24 hours (3 weeks). Stable dose maintenance steroids for central nervous system metastases and low-dose maintenance steroid therapy for other conditions were allowed. The objectives of these dose-escalation trials were to establish the safety, tolerability, and maximum tolerated dose of deforolimus administered as a 30-minute intravenous infusion once weekly (Trial AP23573-02-101)12 or once daily for 5 consecutive days every 2 weeks (Trial AP23573-02-102).13 The 28-day cycle was repeated for patients who tolerated treatment and exhibited tumor response or stable disease. An accelerated titration design and flat-fixed dosing were used. A starting dose of 3 mg was chosen on the basis of preclinical toxicology data. In Table 1. Demographics of the Study Populations

No. of Patients (%) Demographic

Trial 101, n546

Trial 102, n532

15 (32.6) 31 (67.4)

16 (50) 16 (50)

5 (10.9) 40 (87.0) 1 (2.2)

1 (3.1) 30 (93.8) 1 (3.1)

58.3 [24-79]

50.2 [20-71]

9 (19.6) 12 (26.1)

14 (43.8) 18 (56.3)

Sex Women Men

Race Black White Other Mean age [range], y

ECOG performance status 0 1

ECOG indicates Eastern Cooperative Oncology Group.

211

Original Article Table 2. Antitumor Therapy History

No. of Patients (%) Therapy

Trial 101, n546

Trial 102, n532

None Chemotherapy Hormone/immunologic therapy Surgery Radiotherapy

0 (0) 45 (97.8) 12 (26.1)

0 (0) 23 (71.9) 21 (65.6)

31 (67.3) 21 (45.6)

28 (87.5) 16 (50)

Percentage of Patients AE

mIAS (n547)

No mIAS (n524)

100

0

19.2 17 8.5 2.1 2.1 0

8.3 4.2 0 0 0 4.2

61.7 12.8 6.4 4.3 4.3 2.1

29.2 4.2 8.3 0 0 0

44.7 38.3 36.2 4.3 2.1 2.1

45.8 50 41.7 0 4.2 0

70.2 48.9 27.7 23.4 19.2 2.1 2.1

70.8 45.8 16.7 20.8 4.2 4.2 0

Mouth-related AEs

the absence of a grade 2 or worse treatment-related adverse event (AE), the dose was doubled (100%) in the subsequent cohort. If a grade 2 or worse treatment-related AE was observed, then the dose was increased by 50%. If a DLT occurred in Cycle 1, then subsequent doses were increased by 25%. AEs were scored and graded using NCI-CTCAE (version 3) criteria. A list of toxicities was created, and AEs were coded and consolidated from study data. All verbatim AEs were reported by the investigators in the case report forms and were coded using the Medical Dictionary for Regulatory Activities (MedDRA) (version 5.0).14 All primary safety analyses were performed using the MedDRA preferred terms and system organ classes. In addition, several terms were pooled to facilitate the presentation of some AEs. For example, the designation ‘‘mouth sore’’ was used to include the individual MedDRA preferred terms ‘‘aphthous stomatitis, gingival pain, gingival ulceration, glossitis, mouth ulceration, oral discomfort, oral pain, stomatitis, and mucosal inflammation.’’ AEs were stratified according to the presence or absence of concordant oral ulcers and duration. A database was created to further evaluate potential relations between AEs and oral lesions. Because the primary objectives were to describe the clinical course of mTOR-associated oral ulcers and to define their association with other toxicities, no attempt was made to evaluate their relation to deforolimus dose or schedule. Clinical photographs of mucosal ulcerations were obtained by investigators at 2 different study sites.

RESULTS The characteristics of the 2 study populations were similar (Table 1). All patients had received prior treatment for their primary tumors (Table 2). Primary tumor diagnoses varied. The only tumor types with more than 2 patients per diagnosis were renal cell carcinoma (19%), nonsmall

212

Table 3. Incidence of Treatment-Emergent Adverse Events Stratified by Patients With and Without Reported Mammalian Target of Rapamycin Inhibitor-Associated Stomatitis

Mucositis/stomatitis/ mouth sores Pharyngitis Dysgeusia Dysphagia Dry mouth Gingival swelling Pharyngeal edema

Dermatologic AEs Rash NOS Dermatitis acneiform Nail disorder NOS Rash erythematous Acne NOS Dermatitis exfoliative NOS

GI-related AEs Diarrhea Nausea Vomiting Gastroenteritis NOS GI hemorrhage Proctitis

Other AEs Fatigue Anemia Leukopenia Weight loss Neutropenia Infection NOS Sepsis

mIAS indicates mammalian target of rapamycin (mTOR) inhibitor-associated stomatitis; AEs, adverse events; NOS, not otherwise specified; GI, gastrointestinal.

cell lung cancer (9%), mesothelioma (12%), rectal cancer (6%), and colon cancer (5%). Diagnoses with 2 patients per diagnosis (2% each) were breast and bladder cancer, gastrointestinal stromal cell tumor, leiomyosarcoma, and liposarcoma. Thirty-six percent of tumor diagnoses involved only a single patient. Of the 78 patients in the interim dataset, AEs were reported for 71 patients (91%). Mouth ulcers were reported in 47 patients (66%), and these patients were diagnosed with mIAS. The incidence of both mouthrelated and dermatologic-related AEs differed between patients who had mIAS and those who did not (Table 3). The mouth-related AEs that occurred with greater

Cancer

January 1, 2010

mTOR Inhibitor-Associated Stomatitis/Sonis et al

Figure 1. Clustering ulcerations of the floor of the mouth and ventral tongue were noted in a patient who was receiving deforolimus (AP23573). The lesions were ovoid and shallow and had a grayish-white center and a peripheral halo of erythema. The clinical presentation was consistent with that noted in patients with minor aphthous stomatitis (herpetiform type).

frequency in patients with mIAS were dysgeusia (17% vs 4%), pharyngitis (19% vs 8%), and dysphagia (8.5% vs 0%). The dermatologic AEs that occurred with greater frequency in patients with mIAS were nonspecific rashes (62% vs 29%) and acneiform dermatitis (12.8% vs 4.2%). In contrast, the incidence of other gastrointestinalrelated AEs was similar in both patient groups (Table 3). In general, mouth lesions (Figs. 1 and 2) appeared as discrete, ovoid, superficial, well demarcated ulcers with a grayish-white pseudomembrane and measured 1.0 cm in greatest dimension (Figs. 1 and 2). Alternative presentations included a clustered distribution of lesions typical of herpetiform aphthous and larger ulcers similar in appearance to those observed in patients with major aphthous stomatitis. Lesions were confined to the nonkeratinized, movable mucosa, such as the inner aspect of the lips, the ventral and lateral surfaces of the tongue, and the soft palate. Clinical observations demonstrated that ulcers developed acutely, typically peaking in severity within 5 days of deforolimus (AP23573) infusion, and, in the majority of cases, healed spontaneously without scarring in 4 to 5 days. Ulcers largely were accompanied by mouth pain. A higher proportion of mIAS patients than nonMIAS patients were neutropenic (19% vs 4%) and leukopenic (28% vs 17%), but myelosuppression was not a universal finding and was not associated with an increased incidence of infection or sepsis (Table 1). The incidence

Cancer

January 1, 2010

Figure 2. Large ulceration of the lower labial mucosa of a patient receiving deforolimus (AP23573) is shown. The borders were irregular. Although the ulcer was not penetrating, it was extremely painful. The clinical appearance was identical to that of major aphthous stomatitis.

of fatigue and weight loss was similar in mIAS patients and non-mIAS patients.

DISCUSSION Although mTOR inhibitors have been used as immunosuppressive agents in the management of solid organ transplantation and graft-versus-host disease for some time, their use in oncology is relatively new. Ulceration of the oral mucosa reportedly is a frequent and dose-limiting AE arising from the use of mTOR inhibitors in an anticancer setting.15,16 The term ‘‘mucositis’’ has been assigned to these lesions despite the finding that, in widely used criteria for categorizing AEs (such as ICD-9 criteria, MedDRA, and NCI-CTCAE), mucositis tends to be associated with cytotoxic drug and radiation treatment regimens for cancer, whereas mTOR inhibitors are described more accurately as cytostatic. The studies reported herein suggested that the aphthous-like mouth sores that characterize mIAS are distinct from chemotherapy-associated mucositis (Fig. 3) based on their clinical presentation and their association with other cotoxicities. Furthermore, the finding of aphthous-like lesions has been noted consistently in patients who receive mTOR inhibitors for the immunosuppression of solid transplantation rejection.7 The clinical presentation of mIAS closely resembles that of aphthous stomatitis. Lesions appear as distinct, oval ulcers that have a central gray area surrounded by an erythematous band. mIAS ulcers are

213

Original Article

focal or multifocal, well demarcated ulcers that are present in the movable oral mucosa. These self-limiting lesions resemble aphthous oral ulcers and, unlike chemotherapyassociated mucositis, are not associated with other gastrointestinal signs. It is interesting to note that, if the pathogenesis of mIAS is similar to that of aphthous stomatitis, which appears likely, then similar strategies for effective intervention seem reasonable.18 Additional studies are needed to establish factors associated with mIAS incidence, severity, and overall patterns, such as mTORinhibitor dose and schedule.

CONFLICT OF INTEREST DISCLOSURES Figure 3. Ulcerative mucositis associated with traditional cytotoxic chemotherapy is shown. This patient had an extensive, deep ulceration of the ventral tongue. The pseudomembrane covering the ulcer was thicker than that observed on mammalian target of rapamycin inhibitor-associated stomatitis (mIAS) ulcers. Unlike mIAS, an erythematous halo was absent. Typically, the architecture of mucositis-associated ulcers is not defined as well as ulcers associated with mIAS.

localized to the movable mucosa of the mouth and oropharynx and are not present on the more keratinized mucosa of the palate, gingiva, or dorsal surface of the tongue, a finding that distinguishes them from lesions of viral etiology. In general, the lesions of mIAS have a more rapid onset than those of mucositis and resolve in approximately 1 week. Although oral mucositis often occurs concurrent with damage to other areas of the gastrointestinal tract,17 mIAS rarely does so. Among patients who are receiving deforolimus, the incidence of diarrhea, nausea, or vomiting was equivalent in patients with or without mIAS. In contrast, mIAS was associated with the presence of cutaneous lesions: The incidence of nonspecific rash was approximately 3 times higher in patients who experienced mIAS than in those who did not. Although the incidence of neutropenia was higher in patients with mIAS than in patients without mIAS (19% vs 4%), neutropenia was not a ubiquitous finding among individuals with the condition. Nonetheless, the possibility that neutropenia contributes to the etiology of mIAS cannot be disregarded given the similarity between the clinical appearance of mIAS and that of oral lesions observed in neutropenic patients. The results of the current study suggest that mIAS is a distinct entity from conventional chemotherapyinduced or radiotherapy-induced oral mucositis. The oral lesions associated with mTOR inhibitors are discrete,

214

Dr. Demetri has received research support from the following: Novartis, Pfizer, Ariad, Johnson & Johnson, Bristol-Myers Squibb, Infinity Pharmaceuticals, and Daiichi-Sankyo. He has acted as a consultant for the following: Novartis, Pfizer, Ariad, Johnson & Johnson, Genentech, Infinity Pharmaceuticals, ZioPharm, Alnylam, Idera, Bayer, EMD-Serono, Amgen, PamGene, Plexxikon, N-of-One, and Kolltan Pharmaceuticals. He has received honorarium from the following: Novartis and Pfizer. Dr. Demetri has acted as a member of the Scientific Advisory Boards for the following: ZioPharm, PamGene, Plexxikon, and N-of-One, and has acted as Chair of Scientific Advisory Boards for Kolltan Pharmaceuticals. In addition, he has received equity from Plexxikon, N-ofOne, and Kolltan Pharmaceuticals. Dr. Haluska is employed by and has equity in Ariad, Inc. This study was funded by Ariad, Inc.

REFERENCES 1. Baldo P, Cecco S, Gaicomin E, et al. MTOR pathways and mTOR inhibitors as agents for cancer therapy. Curr Cancer Drug Targets. 2008;8:647-665. 2. Figlin RA. Mechanisms of disease: survival benefit of temsirolimus validates a role for mTOR in the management of advanced RCC. Nat Clin Pract Oncol. 2008;5:610-619. 3. Fasolo A, Sessa C. mTOR inhibitors in the treatment of cancer. Expert Opin Investig Drugs. 2008;17:1717-1734. 4. Vignot S, Faivre S, Aquirre D, et al. mTOR-targeted therapy of cancer with rapamycin derivatives. Ann Oncol. 2005; 16:525-527. 5. Scully C, Sonis S, Diaz PD. Oral mucositis. Oral Dis. 2006;12:229-241. 6. Sonis ST, Elting LS, Keefe DM, et al. Perspectives on cancer therapy-induced mucosal injury: pathogenesis, measurement, epidemiology, and consequences for patients. Cancer. 2004;100:1995-2025. 7. Mahe E, Morelon E, Lechaton S, et al. Cutaneous adverse events in renal transplant recipients receiving sirolimus-based therapy. Transplantation. 2005;79:476-482. 8. Scully C. Clinical practice. Aphthous ulceration. N Engl J Med. 2006;355:165-172. 9. Jurge S, Kuffer R, Scully C, et al. Mucosal disease series. Number VI. Recurrent aphthous stomatitis. Oral Dis. 2006;12:1-21. 10. Sonis ST. The pathobiology of mucositis. Nat Rev Cancer. 2004;4:277-284. 11. Porter S, Scully C. Aphthous ulcers (recurrent). Clin Evid (Online). 2005;13:1687-1694. 12. Desai AA, Janisch L, Berk LR, et al. A phase I trial of a novel mTOR inhibitor AP23573 administered weekly

Cancer

January 1, 2010

mTOR Inhibitor-Associated Stomatitis/Sonis et al

(wkly) in patients(pts) with refractory or advanced malignancies: a pharmacokinetic (PK) and pharmacodynamic (PD) analysis [abstract]. J Clin Oncol. 2004;22(14S):3150. 13. Mita MM, Mita AC, Chu QS, et al. Phase I trial of the novel mammalian target of rapamycin inhibitor deforolimus (AP23573; MK-8669) administered daily for 5 days every 2 weeks to patients with advanced malignancies. J Clin Oncol. 2008;26:361-367. 14. Brown EG, Douglas S. Tabulation and analysis of pharmacovigilance data using the Medical Dictionary for Regulatory Activities. Pharmacoepidemiol Drug Saf. 2000;9:479-489. 15. Raymond E, Alexandre J, Faivre S, et al. Safety and pharmacokinetics of escalated doses of weekly intravenous infu-

Cancer

January 1, 2010

sion of CCI-779, a novel mTOR inhibitor, in patients with cancer. J Clin Oncol. 2004;22:2236-2247. 16. Chan S, Scheulen ME, Johnston S, et al. Phase II study of temsirolimus (CCI-779), a novel inhibitor of mTOR, in heavily pretreated patients with locally advanced or metastatic breast cancer. J Clin Oncol. 2005;23:5314-5322. 17. Aprile G, Ramoni M, Keefe D, Sonis S. Application of distance matrices to define associations between acute toxicities in colorectal cancer patients receiving chemotherapy. Cancer. 2008;112:284-292. 18. Chuang P, Langone AJ. Clobetasol ameliorates aphthous ulceration in renal transplant patients on sirolimus. Am J Transplant. 2007;7:714-717.

215