Selection criteria for the subjects were as follows: i) They were histologically or cytologically diagnosed with malignant tumors (without tumor type restriction) and were required to receive chemotherapy; ii) no restrictions were imposed on their chemotherapy plan, which involved the application of chemotherapeutic drugs known to induce moderate or severe vomiting [the severity of chemotherapy-induced nausea was defined according to the NCCN Guidelines^® for Antiemesis (8)]; iii) the patients did not receive any additional treatments, particularly chemotherapeutic drugs, from days 2–7 in a cycle of chemotherapy or antiemetics, tranquilizers, psychostimulants, antihistamines or hormones from days 1–7; iv) their age ranged from 18 to 75 years for either gender and their Karnofsky performance status scores were ≥60; v) their life expectancy was >3 months; vi) their bone marrow was able to produce blood sufficiently; vii) they had normal liver and renal functions; viii) their electrocardiograms were basically normal, which satisfied chemotherapeutic eligibility; ix) they had recovered from toxic reactions (with the exception of baldness and nail changes) induced by previous treatment at least three weeks before the last time of radiotherapy and chemotherapy; and x) they had signed an informed consent form. The study was approved by the ethics committee of Zhejiang Cancer Hospital, Hangzhou, China.

Since granisetron hydrochloride injection, which is currently administered in clinics, is apt to reduce nausea and vomiting, this study adopted a multi-centered, randomly stratified, double-blind, double-dummy, parallel-group and positive-controlled trial method. Based on stratification factors, including the severity of chemotherapy-induced nausea, gender and whether or not this was the first time the patient was receiving chemotherapy, the palonosetron hydrochloride injection (experimental) and granisetron hydrochloride injection (control) groups were created, with 120 patients allocated to each group. Eight centers were included in the study through competition and the central randomization system was applied.

Palonosetron hydrochloride injection was provided by Zhejiang Puluo Kangyu Natural Medicine Co., Ltd. (batch no. 090504), whereas granisetron hydrochloride injection was provided by Ningbo Teampharm Co., Ltd. (batch no. 090301, Ningbo, China). A double-blind and double-dummy method was adopted. The experimental group were administered palonosetron hydrochloride injection (0.25 mg) plus a granisetron hydrochloride injection simulated agent, whereas the control group were administered granisetron hydrochloride injection (3 mg) plus a palonosetron hydrochloride injection stimulating agent. Randomization was performed using SAS software, in accordance with the stratification factors. An emergency letter for unblinding was prepared for each patient. The drugs were diluted with physiological saline to 40 ml and administered to the patients half an hour before chemotherapy. The time taken to administer intravenous injection was >5 min.

The curative effect for vomiting was evaluated according to the following criteria (7,8): i) Complete remission (CR), 0 times/24 h; ii) partial remission (PR), 1 time/24 h; iii) mild remission (MR), 2–5 times/24 h; and iv) failure (F), >5 times/24 h. The remission rates were calculated as follows: CR rate = number of vomiting-free cases/total number of cases; PR rate = number of PR cases/total number of cases; and the effective rate = number of CR+PR+MR cases/ total number of cases. The three rate indices were calculated according to three observation intervals of acute (0–24 h), delayed (24–120 h) and full course (0–120 h) vomiting, respectively.

The effect on nausea was evaluated according to the following criteria: i) CC, normal and nausea-free; and ii) partial control (PC), poor appetite with no changes in food habits (mild nausea) or decreased in food intake, no marked weight loss, dehydration or malnutrition and infusion time of ≤24 h (moderate nausea). The control rates were calculated as follows: CC rate = number of CC cases/total number of cases and PC rate = number of CC+PC cases/total number of cases.

The patients were observed every day during the trial and all adverse events were carefully recorded. Examination of vital signs, electrocardiograms, blood routine, urine routine, liver function, renal function and electrolyte levels were performed prior to and after treatment for safety. Adverse events were evaluated according to the National Cancer Institute CTCAE v3.0 criteria (9,10).

Biostatistical analysis was performed by an independent third party (the Health Statistics Teaching and Research Section of The Second Army Medical University, Shanghai, China) using SAS 10.0 software. All rejected and suspended cases were statistically described. Demographic data and other basic indices were compared between the two groups using the χ² test (or Fisher’s exact probability test), Cochran-Mantel-Haenszel (CMH) test, Student’s t-test and variance analysis or nonparametric statistical analysis methods. Test methods, including logistic regression analysis, the CMH test and covariance analysis, were carried out to the main curative indices for non-inferiority analysis and the non-inferiority threshold was determined to be 15% (Δ=15%). Adverse events in the two groups were statistically described and compared using the χ² test or the Fisher’s exact probability test. Comparisons of vital signs, laboratory examination and electrocardiogram results were based upon the statistical description and analysis of changes observed prior to and after treatment. The mean values and incidence rates before and after drug administration were compared when necessary. α=0.05 was used for all hypothesis tests.

A total of 240 patients were involved in this study and they were evenly divided into the experimental and control groups (n=120). A total of 117 cases in the experimental group and 119 in the control group were enrolled in the full analysis set (FAS), 114 cases in the experimental group and 116 in the control group were enrolled in the per-protocol population set and 118 cases in the experimental group and 120 in the control group were enrolled in the safety analysis data set. The number of suspended cases in the experimental and control groups were 1 and 0 and the number of rejected cases were 7 and 4, respectively; no significant differences were identified. The demographic and baseline characteristics and vital signs of the two groups were similar and no significant differences were identified (Tables I and II).

No significant difference was identified between the acute vomiting CR rates of the experimental and control groups (49.12 vs. 42.24%, respectively; P=0.350; 95% CI, -5.96-19.73; Table III). The non-inferiority test showed that palonosetron hydrochloride injection was not inferior to granisetron hydrochloride injection (P<0.001).

A significant difference was identified between the delayed vomiting CR rates of the two groups (51.75 vs. 31.03%; P=0.002; 95% CI, 8.27–33.17; Table IV). The non-inferiority test showed that palonosetron hydrochloride injection was not inferior to granisetron hydrochloride injection (P<0.001).

No significant difference was identified between the full course vomiting CR rates of the two groups (37.72 vs. 27.59%; P=0.121; 95% CI, -1.92-22.19; Table V). No significant difference was identified between the vomiting control times (7.62±11.68 vs. 9.72±16.01; P=0.573; Table VI), treatment failure times (days; 1.41±1.01 vs. 1.36±0.70; P=0.712; Table VII) or the acute vomiting CR rates (32.46 vs. 27.59%; P=0.136; Table VIII) of the two groups.

Adverse events in this study included any diseases, newly-emerged symptoms, abnormal vital signs or laboratory results and the aggravation of original symptoms or vital sign abnormalities occurring during the clinical drug trial, regardless of whether they were associated with the investigational drugs. Severe adverse events included death, threat to life, permanent or definite disabilities or handicaps, hospitalization or the extension of hospitalization length, congenital disabilities or birth defects, drug overdose and any other severe medical emergencies. Significant adverse events include those listed in adverse events and evident hematological or other laboratory test result abnormalities during drug administration, which may only be cured through targeted treatment.

In the experimental and control groups, 68 (57.63%) and 81 cases (67.50%), respectively, experienced adverse events, 51 (43.22%) and 51 cases (42.50%), respectively, experienced significant adverse events and 0 (0.00%) and 1 case (0.83%), respectively, experienced severe adverse events; no significant differences were identified between the two groups. Adverse reactions in the experimental and control groups were experienced in 17 (14.41%) and 15 cases (12.50%), respectively, and no significant difference was detected between the two groups (Table IX).

Adverse events which occurred at an incidence rate of >10% included leucopenia (11.02% in the experimental group and 22.50% in the control group) and myelosuppression (8.47% in the experimental group and 10.83% in the control group). These events were correlated with the bone marrow functionality of patients, as opposed to with the investigational drugs.

Grade 4 myelosuppression was the one severe adverse event which occurred in the control group, however this was not correlated with the investigational drug and was eventually cured.

Adverse reactions which were associated with the investigational drugs included constipation, asthenia, diarrhea, dizziness and abdominal distention. No statistically significant differences were observed between the two groups. The incidence rates of these reactions in either group were not >10%.