This study was presented at the annual meeting of American Society of Hematology (ASH), held in Orlando, Florida on December 7, 2010.
Conflict of interest: Nothing to report
Correspondence to: Jung Mi Oh, College of Pharmacy, Seoul National University, Seoul, Republic of KoreaDepartment of Internal Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea. E-mail: firstname.lastname@example.org email@example.com
Oral mucositis (OM) is one of the most common and debilitating complications reported by patients undergoing intensive chemotherapy with hematopoietic stem cell transplantation (HSCT). Approximately 40–80% of patients undergoing HSCT suffer from OM with various severities which may subsequently limit tolerability to chemotherapy and adversely affect their daily functioning and quality of life (QOL) [1, 2]. From a patient's perspective, OM is the most serious and detrimental toxicity that can lead to resource-intensive episodes such as severe pain requiring narcotic analgesics and parenteral nutrition, prolonged hospitalization, and potentially life threatening sequelae [3, 4]. Hence, OM has become a major treatment-limiting toxicity and many trials have been targeting toward preventing or treating OM. To present day, there is no known effective modality for the prevention or treatment of OM except palifermin (recombinant human keratinocyte growth factor). Palifermin is approved for the prevention of OM in patients with hematologic malignancies receiving autologous or allogeneic HSCT [5, 6]. Therefore, palifermin along with most of current management is used as the supportive care .
Epidermal growth factor (EGF) was first discovered in the submaxillary gland of a rat in 1962, and is present in various normal tissues and body fluids . EGF plays an important role in maintaining tissue homeostasis, as it regulates epithelial cell proliferation, growth, and migration. Thus, EGF is fundamental in the wound healing and tissue regenerating process [9, 10]. Biosynthetic EGF has been shown to enhance the mucosal wound healing process in the animal model [9, 11], which suggests that it may be effective in treating mucotoxic regimen-induced OM.
Recombinant human EGF (rhEGF, Easyef® topical solution 0.005%, Daewoong Pharmaceutical Company, Seoul, Korea, January 2004) was originally developed and approved for the treatment of diabetic foot ulcer in Korea and several Southeastern Asian countries. It is composed of rhEGF 600,000 I.U. as the active ingredient and methyl parahydroxybenzonate as a solvent. The approved regimen of rhEGF spray is twice daily applications to the wound, and it was not found to be absorbed systemically in a phase I study. Recently, a randomized phase II study illuminated the promising therapeutic efficacy and minimal toxicity of the topical rhEGF solution on radiation-induced OM in head and neck cancer patients . In this study, rhEGF significantly reduced the incidence of severe OM at the primary endpoint (a 64% response in the 50 μg/mL rhEGF group vs. a 37% response in the control group; P =0.025). However, the efficacy and safety of rhEGF have not yet been evaluated in patients receiving intensive chemotherapy for hematologic malignancy.
In this study, we evaluated the effect of topical rhEGF spray for the prevention and treatment of OM in patients receiving intensive chemotherapy followed by HSCT for hematologic malignancies. This study additionally presents the impact of rhEGF on patient self-reported mouth and throat soreness (MTS), as well as the resulting limitations on daily functional activities using the modified version of OM Daily Questionnaire (OMDQ) instrument [13, 14].
Patients and Methods
Patients were eligible for this study if they were at least 18 years of age and were scheduled to receive intensive chemotherapy followed by autologous or allogeneic HSCT between March 2009 and June 2010 for their hematologic malignancy. All patients were required to have normal oral cavity at baseline defined as the National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (CTCAE) or World Health Organization (WHO) scale of grade 0 and an Eastern Cooperative Oncology Group performance status of 0–2 [15, 16]. Exclusion criteria were previous chemotherapy, radiotherapy, or surgery within 3 weeks, or history of hypersensitivity to the study drug or other similar drugs. Patients who participated in clinical trials that could potentially affect the results of this study within 4 weeks or were participating in such a trial at the time of patient selection were also excluded. This study was approved by the Institutional Review Board of the Seoul National University Hospital (IRB Registration No. H-0809-001-255) and was conducted in accordance with the precepts established by the Helsinki Declaration. Signed informed consents were obtained from all the patients before study entry. This trial was registered at www.clinicaltrials.gov as #NCT00845819.
This phase II study is a randomized, double-blind, placebo-controlled trial to evaluate the efficacy and safety of rhEGF for the prevention and treatment of OM in patients receiving intensive chemotherapy undergoing HSCT. Patients were randomly assigned to either the rhEGF group or the placebo group at a 1:1 ratio by the Medical Research Collaborating Center, Seoul National University Hospital, Seoul, Korea. A preplanned interim analysis of this study was conducted on 58 patients who were assigned to receive either rhEGF (N = 29) or placebo (N = 29).
RhEGF and placebo were supplied as multidose spray from the Daewoong Pharmaceutical Company (Seoul, Korea). Patients were instructed to spray either 50 μg/mL rhEGF or placebo over their entire oral mucosa twice daily and to swallow any remaining residue in the oral cavity. Study investigators or nurses checked the appropriate use of spray every day. To confirm proper application of the study spray in each patient, study investigators or nurses also checked the residual volume left in a study spray on a daily basis. RhEGF was administered at the start of intensive chemotherapy until patients recovered from neutropenia (absolute neutrophil count > 1,000 per μL for three consecutive days) and their OM disappeared. Any drugs which could effect on the recovery of OM were not allowed during the study period. For the infection prophylaxis, however, mouthwash with chlorhexidine, povidone iodide, and nystatin was permitted. Hematopoietic growth factors were permitted within the clinically labeled indications.
The primary efficacy endpoint was the incidence of a NCI CTCAE grade ≥2 OM. The secondary endpoints included the incidence and duration of both NCI and WHO grade ≥3 OM and patient-reported QOL according to the sum of area-under-the-curve (AUC) scores from the modified OMDQ (Fig. 1) throughout the study period, the duration of hospitalization, and the incidence and duration of total parenteral nutrition (TPN) and opioid analgesics usage for OM.
A qualified researcher graded the severity of OM using the following two scales: the NCI CTCAE version 4.0 which classifies the severity of OM by its morphology, and the five-grade WHO oral-toxicity scale which also considers its functional disability [15, 16]. OM grading was performed daily throughout the study period. Patient-reported QOL during the study period was also evaluated daily using the modified version of OMDQ. The original questionnaire is composed of 10 questions classified into four categories: overall health, severity of MTS, functional limitations due to MTS, and severity of diarrhea [13, 14]. Of the categories above, the last two questions that were related to diarrhea were excluded from this study (Fig. 1) since the study drug was a topical spray applied locally inside the mouth and throat. A higher score for the overall health endpoints indicated a more favorable response, whereas a higher score for MTS-related endpoints indicated an unsatisfactory response. Information on the use of TPN and opioid analgesics was also collected daily.
Compliance and adverse events
Patient compliance was assessed through daily confirmation of patient-documented administration record. Adverse events related to the topical administration of rhEGF and placebo was graded based on the NCI CTCAE version 4.0 .
Based on the previous study , a sufficient sample size to detect a difference of 27% in the incidence of NCI grade ≥2 OM between two groups with 80% power and a two-sided significance level of 5% was calculated as previously described . In addition, a preplanned interim analysis was considered using the O'Brien-Fleming design with an alpha of 0.003 . A dropout rate of 10% would result in 69 patients for each group to complete the study. Categorical variables were summarized as the value and percentage. Statistical analysis of categorical variables was performed using Pearson's chi-square test or Fisher's exact test, as appropriate. Continuous variables were noted by the median value and range. The Mann–Whitney U test was used to compare non-parametric continuous variables. Post hoc subgroup analysis was used to evaluate outcomes according to the type of conditioning regimen administered.
In the modified version of OMDQ analysis, AUC was calculated from day −8 to day +28 using the trapezoidal method. For each non-missing assessment, the trapezoidal area was calculated as (assessment + next assessment)/[(No. of days in between)/2] . The criteria for a valid AUC calculation were ≥ one assessment between day −8 and day −1; ≥ three assessments between day 0 and day +14; and ≥ one assessment between day +15 and day +28. Any missing assessments took the value of the nearest non-missing assessment for patients who met the AUC calculation criteria. The AUC for the entire study duration was the sum of all the trapezoidal areas.
All of the statistical tests were two-sided, with the significance of the results defined as P < 0.05. Analysis was performed using PASW Statistics version 18.0 (SPSS, Chicago, IL).
Patient demographics and treatment summary
As of June 2010, 58 patients were randomized and enrolled in this study. One patient in each group withdrew his consent at the beginning of the study. Thus, this analysis on the clinical response and safety profiles was performed on 56 patients who received at least one dose of rhEGF (N = 28) or placebo (N = 28).
Baseline demographics and clinical characteristics of the 56 patients are summarized in Table 1. Patient median age in the rhEGF group was 56.5 years (range: 18–63 years) and 51 years (range: 19–65 years) in the placebo group. The number of male patients was 15 (53.6%) in both groups. The majority of patients were diagnosed with multiple myeloma or non-Hodgkin lymphoma. Autologous transplantation was performed on 26 patients (89.7%) in each group, and the rest underwent allogeneic transplantation. Patient characteristics were similar between the autologous and allogeneic HSCT groups. Five and six patients in the rhEGF and placebo group, respectively, received reduced doses of chemotherapy (P = 0.737). All patients had a history of prior chemotherapy, and the number of prior chemotherapy was not significantly different between the two groups (6.0 days [range: 1.0–19.0 days] in the rhEGF group versus 6.0 days [range: 2.0–27.0 days] in the placebo group; P = 0.785). Six patients (21.4%) in the rhEGF group and seven patients (25.0%) in the placebo group had prior radiotherapy. The proportion of patients with hepatic (serum AST or ALT ≥ 40 IU/L) or renal insufficiency (serum creatinine ≥1.5 mg/dL) was similar between the groups: 21.4% in the rhEGF group versus 17.9% in the placebo group, P = 0.737; 17.9% in the rhEGF group versus 14.3% in the placebo group, P = 1.000, respectively. All other baseline demographics and clinical characteristics were similar between the two groups.
The median duration of rhEGF treatment was 19.5 days (range: 15.0–44.0 days) in the rhEGF group and 18.5 days (range: 15.0–61.0 days) in the placebo group (P = 0.453). The median patient compliance rate to the study drug during the course of treatment was 97.2% (range: 44.4–100%) in the rhEGF group and 95.5% (range: 19.4–100%) in the placebo group (P = 0.218).
Incidence and duration of OM
The incidence of NCI grade ≥2 OM was 78.6% in the rhEGF group and 50.0% in the placebo group (P = 0.0496). Patients reached their NCI grade 2 on day 11.0 (median; range: 6.0–21.0 days) in the rhEGF group and on day 10.0 (median; range: 7.0–19.0 days) in the placebo group (P = 0.843). The median duration of NCI grade ≥2 OM among patients with adverse effects was 8.5 days (range: 1.0–25.0 days) in the rhEGF group and 14.5 days (range: 1.0–36.0 days) in the placebo group (P = 0.262).
Similar results were observed for other grades of OM as summarized in Table 2. The incidence of NCI grade ≥3 OM was 39.3% in the rhEGF group and 32.1% in the placebo group (P = 0.577), and its median duration was shorter among the 11 rhEGF recipients than the nine placebo recipients (8.0 days [range: 2.0–23.0 days] vs. 16.0 days [range: 1.0–32.0 days]; P = 0.381). The incidence of WHO grade ≥3 OM was 46.4% in the rhEGF group and 28.6% in the placebo group (P = 0.168). The median duration of WHO grade ≥3 OM was 8.0 days (range: 2.0–23.0 days) in the rhEGF group and 18.5 days (range: 4.0–34.0 days) in the placebo group (P = 0.108).
Table 2. Effects of rhEGF on the Incidence and Duration of OM
RhEGF (N = 28)
Placebo (N = 28)
Abbreviations: rhEGF, recombinant human epidermal growth factor; NCI, the National Cancer Institute; WHO, the World Health Organization; OM, oral mucositis.
OM of NCI or WHO grade ≥2
Incidence, No. (%)
In patients with OM of NCI grade ≥2
OM of NCI grade ≥3
Incidence, No. (%)
In patients with OM of NCI grade ≥3
OM of WHO grade ≥3
Incidence, No. (%)
In patients with OM of NCI grade ≥3
In a subgroup analysis of patients receiving high-dose melphalan conditioning regimen (N = 19 in the rhEGF group, N = 16 in the placebo group), the incidence of NCI grade ≥2 OM was 73.7% in the rhEGF group and 37.5% in the placebo group (P = 0.031). However, when exclusively focusing on the NEAM regimen (N = 7 in the rhEGF group, N = 10 in the placebo group), the incidence of NCI grade ≥2 OM was not different between the two groups (85.7% in the rhEGF group, 66.7% in the placebo group, P = 0.585). In contrast, the median duration of WHO grade ≥3 OM tended to be shorter in the rhEGF group (8.0 days [range: 2.0–14.0 days]) than in the placebo group (16.0 days [range: 4.0–23.0 days]; P = 0.057) in this subgroup analysis. Other outcomes in subgroup analyses showed no significant differences between the rhEGF and placebo groups.
Impact of rhEGF on patient-reported QOL
Among all the patients, the score for overall health was higher in the rhEGF group than in the placebo group, with a borderline significance (median AUC, 254.25 [range: 121.50–308.50] vs. 217.25 [range: 70.00–353.00]; P = 0.074). For other MTS and MTS-related scores, patients in the placebo group showed results that were similar with those in the rhEGF group.
In patients with WHO grade ≥3 OM, however, rhEGF significantly reduced limitations in swallowing (median AUC, 26.0 [range: 8.0–75.0] vs. 51.5 [range: 19.0–92.0]; P = 0.039) and drinking (median AUC, 25.0 [range: 7.0–73.5] vs. 55.5 [range: 20.0–86.0]; P = 0.042) compared to placebo (Fig. 2). Although the rhEGF group also showed more favorable responses to other questions, the differences were not statistically significant. The results of the OMDQ analysis of patients with WHO grade ≥3 OM are summarized in Table 3.
Table 3. Effects of rhEGF on OMDQ Questionnaire Among Patients with WHO Grade ≥3 OM
rhEGF (N = 13)
Placebo (N = 8)
Abbreviations: AUC, area under the curve; rhEGF, recombinant human epidermal growth factor; MTS, mouth and throat soreness; OM, oral mucositis.
Q1. Overall health
Q3. MTS-related limitations in daily activities
Q4. Overall MTS
Impact of rhEGF on the administration of TPN and opioid analgesics usage, and hospital stay
Impacts of rhEGF on the other secondary endpoints are summarized in Table 4. Among the patients who were analyzed, 13 (46.4%) required TPN administration during their treatment in the rhEGF group, and 12 (42.9%) in the placebo group (P = 0.788). The median duration of TPN administration was not significantly different between the two groups (0 day [range: 0–25.0 days] vs. 0 day [range: 0–32.0 days]; P = 0.794). Opioid analgesics were administered to 14 patients (50%) in the rhEGF group and 15 patients (53.6%) in the placebo group (P = 0.789). Recipients of rhEGF tended to require less opioid analgesics for OM than placebo recipients, as measured by the median daily dose (0.5 mg/day of morphine equivalents [range: 0–112.8] vs. 17.3 mg/day of morphine equivalents [range: 0–129.0], P = 0.754) and the median duration of morphine administration (0.5 day [range: 0–24.0 days] vs. 2 days [range: 0–48.0 days]; P = 0.492). The difference, however, was statistically insignificant. The median time to neutrophil recovery (absolute neutrophil count ≥500 per μL) was 10.5 days (range: 8–22 days): 10 days (range: 8–22 days) in the rhEGF group and 11 days (range: 8–22 days) in the placebo group (P = 0.212). The incidence of microbiologically documented sepsis was 25.9% (15 patients): 28.6% (eight patients) in the rhEGF group and 25.0% (seven patients) in the placebo group (P = 1.000). No transplant-related mortality was reported. The median follow-up duration was 18.6 months (range: 3.9–25.9 months). The 1-year and 2-year survival rates were 94.2 and 89.9%, respectively.
Table 4. Effects of rhEGF on the Administration of TPN and Opioid Analgesics, and Hospital Stay
Abbreviations: TPN, total parenteral nutrition; ME, morphine equivalent; NCI, the National Cancer Institute; WHO, the World Health Organization; OM, oral mucositis.
Incidence, No. (%)
Opioid analgesic use
Incidence, No. (%)
ME dose, mg/day
Among patients with OM of NCI or WHO grade ≥2
Incidence, No. (%)
Opioid analgesic use
Incidence, No. (%)
ME dose, mg/day
Among patients with OM of WHO grade ≥3
Incidence, No. (%)
Opioid analgesic use
Incidence, No. (%)
ME dose, mg/day
Among the patients with WHO grade ≥3, the median duration of TPN administration was significantly shorter in the rhEGF group (7.0 days [range: 0–25.0 days]) than in the placebo group (16.5 days [range: 10.0–32.0 days]; P = 0.012). A similar result was observed for the median duration of opioid analgesics use for pain induced by OM (5.0 days [0–24.0 days] vs. 20.5 days [range, 8.0–48.0 days]; P = 0.018). However, the median daily dose of opioid analgesics administered was not statistically different between the two groups (18.8 mg/day of morphine equivalents [range: 0–112.8] vs. 45.3 mg/day of morphine equivalents [range: 14.6–129.0]; P = 0.167). The rhEGF group showed a significantly shorter length of hospital stay than the placebo group (27.0 days [range: 18.0–65.0 days] vs. 41.0 days [range: 27.0–90.0 days]; P = 0.047).
The rhEGF treatment-related adverse events
Adverse events related to topical administration showed no difference between the two groups. The adverse events experienced in the rhEGF group were nausea (N = 2, 6.9%), oral pain (N = 1, 3.4%), and taste alteration (N = 1, 3.4%). All of these events were mild in severity and transient. No grade 3 or 4 adverse events were observed during the treatment course.
To the best of our knowledge, this study is the first randomized, placebo-controlled, phase II clinical trial that evaluated the efficacy and safety of rhEGF on OM induced by intensive chemotherapy with HSCT for hematologic malignancies. The most apparent one is the proportion of patients with OM of NCI grade ≥2 was higher in the rhEGF group than in the placebo group, with a borderline significance. However, the incidence of WHO or NCI grade ≥3 OM was similar between the two groups. In contrast, the duration of OM in patients with NCI grade ≥2 tended to be shorter in the rhEGF group than the placebo group. This result was reproduced when the analysis was performed on other grades of OM.
Issues of patient QOL and related complications after the development of chemotherapy-induced OM should be considered since OM is one of the most debilitating adverse effects that affect QOL and ultimately increases the mortality . For this reason, patients with WHO grade ≥3 OM who generally have severe pain and limitations in oral intake were further analyzed. In these patients, the results of the OMDQ analysis showed that rhEGF significantly reduced limitations in swallowing and drinking. Consistent with the impact of rhEGF shown from the OMDQ, the duration of TPN and opioid analgesics use was significantly shorter in the rhEGF group than in the placebo group. The length of hospital stay was also significantly shorter in the rhEGF group than in the placebo group. Since the patients with severe OM in the rhEGF group showed better results in terms of resource-intensive episodes than in the placebo group, there is a need for further economic analysis after the completion of this study, as was seen in the case of a previous economic analysis of palifermin .
The results of the primary endpoint of this analysis showed a discrepancy from the results of a previous study that showed a promising efficacy of rhEGF in the prevention of OM in patients undergoing radiotherapy with or without chemotherapy for head and neck cancer . This discrepancy could possibly be explained by the difference in mucosal response between chemotherapy and radiotherapy. Given that OM is directly related to the treatment-related factors, its time course and severity of OM vary with factors such as the type of chemotherapy or radiotherapy, total dosage, fractionation, and duration of treatment . A difference in the frequency of OM assessment between these two studies (weekly in the previous study vs. daily in our study) could also cause this discrepancy. Additionally, we could find several clues explaining this discrepancy from a previous literature. A previous animal study demonstrated that EGF increased the oral mucosal breakdown of Golden Syrian hamsters in the face of antineoplastic therapy using 5-fluorouracil . In this animal study, the course and extent of mucositis was influenced by the timing of EGF administration in the course of chemotherapy. The results of this animal study suggested that EGF administration during chemotherapy could increase mucosal sensitivity to cancer chemotherapy through the accelerated buccal mucosal cell division. It is known that cancer therapy affects rapidly dividing cells and that the normal oral basal epithelial cell turnover rate is 7–14 days . The rhEGF acting as a growth factor could have increased the turnover rate of oral basal epithelial cells, thus theoretically, it is possible that the rhEGF administered during chemotherapy could have negatively stimulated the development of OM. In contrast, one can speculate that rhEGF performed its full epidermal regeneration activity when it was administered during the recovery period following high-dose chemotherapy. Chemotherapeutic agents administered as part of the conditioning regimen in this study have different elimination half-lives that range from 60 min to 75 hr [23-26]. In most cases, accordingly to first-order kinetics, it takes approximately three to five half-lives for a drug to be essentially eliminated from the body . In our analysis, the duration of OM actually tended to be shorter in the rhEGF group as compared to the placebo group. These data suggest that rhEGF could have mucosal regeneration activity when it was administered during the recovery period following high-dose chemotherapy—after elimination of chemotherapeutic agents from the body. Thus, these results warrant further clinical trials using a different administration schedule of rhEGF in the setting of HSCT.
In conclusion, the rhEGF did not reduce the incidence of NCI grade ≥2 OM. The patients with severe OM with WHO grade ≥3 in the rhEGF group, however, showed better results than those in the placebo group for several secondary endpoints. RhEGF was generally well tolerated in most patients with a favorable safety profile. Unlike palifermin which should be administered intravenously before and after HSCT, rhEGF is an oral spray that can be topically applied to patients with severe OM irrespective of any prior history of palifermin therapy or cryotherapy. This study provides a rationale for continual exploration on the effects of rhEGF for the prevention and treatment of chemotherapy-induced OM.
The authors acknowledge the efforts of Eun-hee Park R.N. for collecting clinical data and providing administrative supports; Kyung-Sang Yu, M.D., Ph.D. for providing administrative supports; the Medical Research Collaborating Center, Seoul National University Hospital, for statistical consultations; and Daewoong® Pharmaceutical Company, for supplying study drugs.