This article is mentioned in:

Introduction

Multiple myeloma (MM) is a malignant neoplasm characterized by the clonal proliferation of plasma cells in the bone marrow that produce monoclonal protein (M protein) (1,2). Although MM remains an incurable malignancy, the survival rates have improved markedly (2). The first notable improvement came following the introduction of autologous stem cell transplantation (3,4). In addition, the use of drugs with antiangiogenic, antiproliferative and immunomodulatory effects, including thalidomide (5) and lenalidomide (6,7), and proteasome inhibitors, including bortezomib (8,9), were introduced (10). These agents have been combined with melphalan, resulting in high response rates in patients with MM (11–14).

Melphalan is a cytotoxic agent used extensively in the treatment of MM (15,16). For decades, the combination of melphalan with prednisolone provided the standard of care for elderly patients with MM. Previously, a bortezomib plus oral melphalan and prednisolone (VMP) regimen was demonstrated to exert synergistic anticancer effects (17,18), and thus this has become the current standard regimen for patients with MM (11).

The oral bioavailability of melphalan is known to be widely variable (19–21). For example, Sviland et al (22) demonstrated that pretreatment of the patients with the histamine 2 (H2) receptor blocker, cimetidine, reduced the oral bioavailability of mephalan by ~35%. This reduction is thought to stem from the poor absorption of melphalan (19), since its solubility is known to decrease under alkaline pH conditions (23,24). However, this pharmacokinetic interaction has not been conclusively demonstrated to be associated with any reduced clinical efficacy. Furthermore, there is no mention of any drug interactions between oral melphalan and gastric antisecretory drugs in the medical package insert in Japan.

The aim of the present study was to assess whether the clinical efficacy and toxicity of melphalan were influenced by the concomitant use of gastric antisecretory drugs in patients with MM receiving VMP therapy. The clinical significance of the interaction between melphalan and gastric antisecretory drugs was also discussed.

Patients and methods

Ethics statement

The present study was reviewed and approved by the Institutional Review Boards of the Japan Community Health care Organization Kyoto Kuramaguchi Medical Center (Kyoto, Japan; IRB no.: 2015012602). Written patient consent was waived due to the retrospective and observational nature of the study.

Study population and design

A retrospective study was performed at the Japan Community Health care Organization Kyoto Kuramaguchi Medical Center (Kyoto, Japan). A total of 12 patients with MM who received VMP therapy between December 2011 and November 2014 were included. However, two patients from this cohort were excluded from the analysis, due to an inability to measure the level of M protein throughout the first cycle of treatment. The remaining 10 patients were divided into two groups: The concomitant (three patients) and control (seven patients) groups, according to the additional use of gastric antisecretory drugs, or the lack thereof. The VMP regimen consisted of bortezomib (1.3 mg/m2) administered on days 1, 8, 15 and 22, and oral melphalan (6 mg/m2) and prednisolone (40 mg/m2) administered on days 1–4 per one cycle (35 days).

Data collection and definitions

Two parameters were used as measures of clinical efficacy: (i) a reduction in the level of M protein; and (ii) the best response according to the International Myeloma Working Group criteria (25) throughout the course of treatment. The hematological and gastrointestinal (GI) toxicity of melphalan was assessed according to the Common Terminology Criteria for Adverse Events version 4.0 (26), using the highest grade available throughout the duration of the three-cycle treatment for the analysis.

Statistical analysis

The data are expressed as the mean ± standard deviation, or the median (range). Comparisons of the two groups were performed using the unpaired Student's t-test (parametric) or the Mann-Whitney U test (non-parametric). Differences between observed and expected frequencies were evaluated using Fisher's exact probability test. P<0.05 was considered to indicate a statistically significant difference.

Results

Characteristics of the patients

Table I shows the characteristics of the 10 patients enrolled in the present study. Seven patients were not co-administered gastric antisecretory drugs (control group), and three patients were co-administered gastric antisecretory drugs (two patients were administered rabeprazole sodium, one patient was administered famotidine) in the VMP regimen (concomitant group). Patient no. 1 was changed to the VMP regimen after having received oral melphalan and prednisolone therapy during cycle 1. No significant differences were observed in the characteristics of the patients between the control and concomitant groups.

Table I. Characteristics of the patients.

Table I.

Characteristics of the patients.

	Control							Concomitant
Patient no.	1	2	3	4	5	6	7	8	9	10	P-value
Age (years)	79	80	71	74	78	80	63	69	72	70	0.137a
Sex	F	M	M	M	F	F	F	M	F	F	1.000b
Myeloma type	IgG	IgG	IgG	IgG	BJP	IgG	IgG	IgG	IgG	IgG	1.000b
ISS stage	II	III	II	I	II	III	I	III	III	II	0.888b
Previous treatment (Number)	Yes (1)	No (0)	No (0)	No (0)	No (0)	No (0)	No (0)	No (0)	No (0)	No (0)	1.000b

a Mann-whitney U test,

b Fisher's exact probability test. BJP, Bence Jones protein; IgG, immunoglobulin G; ISS, international staging system.

Table II shows the doses of chemotherapeutic agents in the VMP regimen in the control and concomitant groups. The doses of melphalan were 4.6 and 5.5 mg/m2 in the control and concomitant groups, respectively; the difference between the two doses was not statistically significant. Furthermore, the doses of other components used in the VMP regimen (bortezomib and prednisolone) revealed no statistically significant differences between the control and concomitant groups.

Table II. Doses of the chemotherapeutic agents in the VMP regimen.

Table II.

Doses of the chemotherapeutic agents in the VMP regimen.

	Control (n=7)	Concomitant (n=3)	P-valuea
Bortezomib (mg/m2)	1.3 (1.0–1.3)	1.3 (1.2–1.3)	0.569
Oral melphalan (mg/m2)	4.6 (3.4–8.6)	5.5 (5.5–6.1)	0.210
Prednisolone (mg/m2)	30.0 (12.7–57.5)	37.5 (27.7–38.0)	0.909

{ label (or @symbol) needed for fn[@id='tfn3-mco-0-0-683'] } Data are shown as the median (range).

a Mann-whitney U test.

Clinical efficacy

The levels of M protein over the duration of the treatment period for the two groups are shown in Fig. 1. The levels of M protein in the control group markedly decreased, depending on the number of cycles; however, the levels in the concomitant group remained largely unchanged. The levels of M protein in cycles 2 and 3 were significantly higher in the concomitant group compared with the control group (P<0.01).

Figure 1. Changes in the level of the M protein in the VMP regimen with or without gastric antisecretory drugs. The level of the M protein prior to the start of the VMP regimen was expressed as 100%. Open and closed circles represent the control (without gastric antisecretory drugs) and concomitant (with gastric antisecretory drugs) groups, respectively. Each symbol represents the mean ± standard deviation. **P<0.01, vs. control (unpaired Student's t test).

The other index of clinical efficacy, i.e. the best response in the three cycles, is summarized in Table III. All the patients in the control group were graded as having a partial response (PR) to the drug therapy, whereas all the patients in the concomitant group were graded as having stable disease (SD). The best response to the VMP regimen was significantly higher for the control group than for the concomitant group.

Table III. Best response obtained in three cycles of the VMP regimen.

Table III.

Best response obtained in three cycles of the VMP regimen.

	Response
	PR	SD	P-valuea
Control (n=7)	7	0	0.008
Concomitant (n=3)	0	3

{ label (or @symbol) needed for fn[@id='tfn5-mco-0-0-683'] } PR, partial response; SD, stable disease. The clinical efficacy of the VMP regimen attained its best response in the first three cycles, according to the International Myeloma Working Group criteria.

a Fisher's exact probability test.

Hematological and GI toxicity

Table IV shows the grade and frequency of selected hematological and GI toxicities. Throughout each treatment cycle, the grade and frequency of leukopenia, anemia and thrombocytopenia were comparable in the two groups. Furthermore, no significant differences were observed in the white blood cell count, hemoglobin level, or the platelet count between the two groups (Fig. 2). Grade 1 GI toxicities were observed in the control group, manifesting as vomiting in one patient, and diarrhea and loss of appetite in two other patients. No GI toxicity was observed in the concomitant group.

Figure 2. Changes in the WBC count, Hb level and the PLT count in the VMP regimen with or without gastric antisecretory drugs. The WBC count, Hb level or PLT count before the start of the VMP regimen is expressed as 100%. The open and closed columns show the control (without gastric antisecretory drugs) and concomitant (with gastric antisecretory drugs) groups, respectively. Each bar represents the mean ± standard deviation. NS, not significant (unpaired Student's t-test); WBC, white blood cell; Hb, hemoglobin; PLT, platelet.

Table IV. Hematological and gastrointestinal toxicity.

Table IV.

Hematological and gastrointestinal toxicity.

	Grade	Control (n=7)	Concomitant (n=3)	P-valuea
Leukopenia	Grade 1/2	5	1	0.500
	Grade 3/4	2	2
Anemia	Grade 1/2	5	1	0.500
	Grade 3/4	2	2
Thrombocytopenia	Grade 1/2	3	3	1.000
	Grade 3/4	1	0
Vomit	Grade 1	1	0	–
Diarrhea	Grade 1	2	0	–
Appetite loss	Grade 1	2	0	–

{ label (or @symbol) needed for fn[@id='tfn7-mco-0-0-683'] } Evaluation of hematological toxicity adopted a high grade in the first three cycles, according to the Common Terminology Criteria for Adverse Events, version 4.0.

a Fisher's exact probability test.

Discussion

Limitations to the present study included its retrospective small sample size and a lack of plasma melphalan concentrations. However, the results in the present study are sufficient to call attention to the interaction between melphalan and gastric antisecretory drugs.

In the present study, three of the ten patients received gastric antisecretory drugs in addition to VMP therapy. Of the three patients in the concomitant group, two patients received rabeprazole sodium and one received famotidine. With the exception of the use of gastric antisecretory drugs, the control and the concomitant groups were comparable in terms of the characteristics of the patients (Table I) and the doses of chemotherapeutic agents in the VMP regimen (Table II). However, the decrease in the level of the M protein in treatment cycles 2 and 3 was significantly greater in the control group (Fig. 1), indicating a more favorable clinical response in that cohort. In addition, all the patients in the control group were graded as having a PR to the drug therapy, compared with the SD identified in the concomitant group (Table III). Taken together, these findings suggested that the clinical efficacy of the VMP regimen against MM decreases on co-administration of the gastric antisecretory drugs. As for bortezomib, the pharmacokinetics, efficacy and safety of bortezomib were not affected by co-administration of either the proton-pump inhibitor, omeprazole (27), or the H2 receptor blocker, lafutidine (28).

The hallmark adverse events of melphalan are known to be hematological and GI toxicities. The present study disclosed no significant difference in the analyses of hematological toxicity between the control and concomitant groups (Fig. 2 and Table IV), in stark contrast with the clinical efficacy findings reported in the present study. These results are in agreement with previous reports, which cited no significant differences in hematological toxicity among patients receiving oral melphalan following cimetidine pretreatment (22). Therefore, in spite of the decreased bioavailability of melphalan in the concomitant group, the plasma concentrations were likely to have remained within a range in which myelosuppression would be predicted to occur. The observed differences in GI toxicity (Table IV) may be partly explained by the poorer solubility of melphalan tablets with increasing gastric pH.

Taken together, these results suggested that the interaction of the VMP regimen with gastric antisecretory drugs resulted in a marked decrease in the clinical efficacy of melphalan. As such, oral melphalan should not be co-administered with gastric antisecretory drugs, and this drug interaction should prominently feature in the medical package insert.

The mechanism of the interaction between melphalan and gastric antisecretory drugs remains to be fully elucidated. For example, melphalan was reported to be unstable at alkaline pH, and the dissolution of melphalan tablets occurred more slowly at an increased gastric pH (23). Furthermore, Sviland et al (22) reported a 35.5% decrease in the oral bioavailability of melphalan in patients pretreated with cimetidine. Therefore, drugs that induce a potent and long-lasting inhibition of gastric acid secretion [a pharmacological hallmark of rabeprazole sodium (29,30) and famotidine] would predispose a patient towards this interaction. Food intake was also reported to decrease the absorption rate of oral melphalan (31,32). It is notable that melphalan was administered before the meal in the present study.

Taken together, the results in the present study suggest that this drug interaction may arise from the poor solubility of the melphalan tablets, brought about by an increase in gastric pH caused by the concomitant use of gastric antisecretory drugs, thereby resulting in a diminished clinical efficacy of melphalan.

In conclusion, the interaction between oral melphalan and gastric antisecretory drugs may result in a marked decrease in its clinical efficacy in the treatment of MM. This drug interaction possibly results from a reduction in the absorption of melphalan caused by an increase in gastric pH. Due to the aforementioned limitations of the current study, further research is necessary to obtain definitive evidence for these conclusions.

References