Patients with MM, aged 15–70 years, who achieved VGPR or a better response after auto-PBSCT, were enrolled in this trial. The main inclusion criteria were a good performance status (0–2), an absence of renal dysfunction as assessed by creatinine clearance >60 ml/min, an absence of deep vein thrombosis or pulmonary thromboembolism, and recovery from myelosuppression following auto-PBSCT. Female patients were either postmenopausal, surgically sterilized, or willing to use an acceptable method of birth control for the entire duration of the study. Likewise, male patients consented to using an acceptable method of contraception for the entire duration of the study. The main exclusion criteria included grade ≥2 peripheral neuropathy, active infection, positive serology for human immunodeficiency virus or hepatitis B or C viruses, pregnancy and breast-feeding. This trial was approved by the institutional ethics committee at each participating institution and registered as UMIN000012700. Written informed consent was obtained from all the patients prior to registration in accordance with the Declaration of Helsinki.

The patients required LEN maintenance therapy within 6 months of undergoing auto-PBSCT. The three selected dose levels of LEN (level 0, 5 mg; level 1, 10 mg; and level 2, 15 mg) were determined based on previously published data (6–8). The assigned dose of LEN was administered orally once per day for 12 weeks, without any drug holidays. The patients were able to continue receiving LEN during the scheduled period, unless a dose-limiting toxicity (DLT) occurred, obvious evidence of disease progression appeared, or the patient refused to continue treatment.

According to the National Cancer Institute Common Toxicity Criteria version 4.0 (http://ctep.cancer.gov/protocolDevelopment/electronic_applications/ctc.htm#ctc_40), any hematological or non-hematological adverse event (AE) of grade ≥3, or discontinuation of drug administration due to any grade of AEs directly attributable to LEN, were considered as DLTs. The assessment of disease response was based on the International Uniform Response Criteria (9).

The FID was estimated using the continual reassessment method (CRM) (10,11). Dose escalations or de-escalations for consecutive patient cohorts and the size of each cohort were based on the dose-finding CRM algorithm and clinical judgment. Jumping from level 0 to 2 was not permissible in the CRM algorithm. The target completion rate of maintenance therapy was set at 70%. The FID of this trial was determined as the dose that was closest to the level at which 70% of patients would complete 12 weeks of maintenance therapy without experiencing a severe AE. We determined the starting dose in this trial to be level 1, and the first 3 patients were treated at this level. According to the pre-specified dose de-escalation rule, if DLT was observed in 2 of the first 3 patients, another 3 patients allocated to the second cohort were treated at level 0. This phase I study was expected to require a project sample-size of ≤15 patients.

Between August, 2011 and April, 2014, a total of 11 patients (8 in CR and 3 in VGPR) from five transplant centers in Japan were enrolled in this study, with a median of 99 days (range, 60–168 days) after auto-PBSCT. Using the CRM algorithm, 6 patients were assigned to level 0 and 5 were assigned to level 1 (Fig. 1). the characteristics of these 11 patients are summarized in Table I. The median age was 58 years (range, 41–65 years), and 7 patients were men (64%). The subtypes of monoclonal immunoglobulin (Ig) detected in the 11 patients were IgG (n=4), IgA (n=4), IgD (n=1) and light-chain-only disease (n=2). All the patients received a BOR-containing induction regimen prior to auto-PBSCT; however, no patients received prior THAL.

The time-dependent changes in hematological and non-hematological AEs observed during maintenance therapy with LEN are detailed in Table II. Of the 6 patients assigned to level 0, 3 (50%) experienced neutropenia (grade ≤2) at approximately 8 weeks after initiating LEN maintenance therapy; however, all the patients spontaneously recovered from neutropenia despite continuous administration of LEN. As for the remaining patients, no serious AEs (SAEs) were observed at level 0. As shown in Table II, more patients experienced AEs at level 1. One patient [unique patient number (UPN) 3]developed pneumonia as a consequence of neutropenia in week 10 of treatment; therefore, LEN administration was discontinued at 11 weeks. Another patient (UPN 2) developed both neutropenia and thrombocytopenia in addition to peripheral neuropathy in week 4 of treatment; although this patient's dose was reduced to 5 mg (level 0), thrombocytopenia progressed and LEN administration was discontinued soon after. Another patient (UPN 9) developed liver dysfunction in week 2 of the treatment, which had progressed to grade 2 by week 6. Administration of LEN was discontinued for this patient as well. Grade 3/4 toxicities were not observed in any of the study patients.

The dose-finding CRM algorithm and clinical judgment used in this study are shown in Fig. 1. of the first 3 patients assigned to level 1, 2 patients (UPN 2 and 3) encountered DLT [discontinuation of LEN with the attending physician's judgment based on progressive grade 2 hematological (n=1) and non-hematological (n=1) AEs]. According to the pre-specified dose de-escalation rules for this trial, the dose level for the second patient cohort was de-escalated to level 0. However, none of the 3 patients (UPN 4, 5 and 6) at level 0 in the second cohort experienced any DLTs. In the third cohort, 1 patient (UPN 7) remained assigned to level 0, but did not experience any DLTs. Therefore, the next 2 patients (UPN 8 and 9) were assigned to level 1 in the third cohort, although 1 (UPN 9) of the 2 patients again experienced a DLT (discontinuation of LEN administration occurred with the attending physician's judgment based on a progressive grade 2 non-hematological AE). Consequently, the estimated level 1 completion rate was considerably lower compared with the target level, and the next patient cohort was treated at level 0. Of the subsequent 2 patients (UPN 10 and 11) treated at level 0, as the fourth cohort, neither experienced DLTs. Based on the results of the sensitivity analysis and the observed toxicities at each dose level, we concluded that LEN maintenance therapy at level 0 was acceptable in terms of safety.

The disease status prior to and following LEN maintenance therapy is indicated in Table III. The results from the 8 assessable patients who completed 12 weeks of LEN maintenance therapy revealed no disease progression or obvious response. The 2 patients with VGPR maintained that status and did not achieve CR. The 6 patients with CR also maintained stable disease conditions during LEN maintenance therapy.