1.  INTRODUCTION

Chemotherapy-induced nausea and vomiting (CINV) is one of the factors significantly affecting the patient’s quality of life and reduces compliance to medication. When  antiemetic prophylaxis is not given, the incidence of CINV is as high as 90% with highly emetogenic regimens(HEC) and  30-90% with Moderately Emetogenic regimens (MEC) (1). With extensive research work in this area, control of vomiting is better than nausea control (2). CINV divided into 3 phases: acute phase – on day of chemotherapy (0-24 hrs), delayed phase (25-120hrs) following chemotherapy, and anticipatory phase. The acute phase mediated by serotonin through 5- HT3 receptor, whereas the delayed phase is mediated by  substance P through NK 1 receptor (3). The chemotherapy regimens are classified into 4 levels according to their emetogenic potential: Highly emetogenic ->90%, moderately emetogenic - 30 % to 90%, low-10% to 30%, minimal-<10% (1).

At present, 5-HT 3 receptor antagonists, dexamethasone, NK 1 receptor antagonists are used for the antiemetic prophylaxis of CINV. But still, patients experience nausea and vomiting, especially nausea after chemotherapy. Studies assessing the patient’s quality of life using the Functional Living Index Emesis (FLIE) questionnaire signifies that CINV had a negative impact on the Quality of life (QOL) of the patient (7). The cost of NK receptor antagonist is also high and creates a cost burden on a patient in a low socioeconomic status (8). So, an alternative cheaper agent is needed.  Olanzapine, an atypical antipsychotic, is used as an antiemetic agent through its ability  to block the multiple neurotransmitters in the brain such as dopamine, histamine, serotonin, acetylcholine (9). It is recently added to guidelines (National Comprehensive Cancer Network) as one of the components of a four-drug combination to manage the CINV due to highly emetogenic chemotherapeutic agents. Olanzapine (10 mg) is also recommended for use in CINV due to moderately emetogenic agents (1). However, recent studies have suggested using  a low dose (5mg) to reduce sedation with the same efficacy in patients undergoing cancer chemotherapy (10). Pregabalin, a gabapentinoid, acts on the alpha 2 delta subunit of the presynaptic calcium channel and inhibits the release of many excitatory neurotransmitters, including substance P, dopamine, serotonin, glutamate and norepinephrine(11). A meta-analysis had shown that pregabalin reduced the incidence of postoperative nausea and vomiting (12). But, there is no evidence suggesting the role of pregabalin in the management of CINV. However, it is observed that gabapentin, which is another gabapentinoid, has shown some efficacy in the reduction of CINV(13). We theoretically consider that there might be a synergism as one drug (pregabalin) will reduce the release of neurotransmitters, and another drug (olanzapine) will block the action of the released neurotransmitters. So in this study, we will check the role of pregabalin (75 mg) combined with the olanzapine 5mg.

Sample size calculation

A study done by Wu et al(14)  showed the improvement in the proportion of patients in terms of no nausea was 20% in the olanzapine group patients receiving MEC regimen, 35% with of the patients in the control group experiencing no nausea. In this study, we are expecting an increase in the effect size proportion to 30% due to the addition of pregabalin along with olanzapine. So, a sample size of N=84 (42 in each arm) with alpha as 0.05 and 80% power are calculated.

We are planning for four analyses (3 interim and the final analyses) during the study. For the interim analysis to control the type 1 error inflation, the O’Brien-Fleming boundary is defined as provided in figure 1. The sample size provided in the figure will be used only if the allocation ratio remains 1 throughout the study. The study will be terminated if the z value of the effect size crosses the boundary on either side during an interim analysis. However, to minimise the exposure to harmful or ineffective treatment to the trial participants, adaptive randomisation will be adopted after each interim analysis. This procedure will change the allocation ratio (discussed in randomisation in detail) after each interim analysis. Because of interim analysis and adaptive randomisation, the sample size will change during the course of the study based on the response of the trial participants after the first interim analysis

 Normal test statistics at bounds 

Figure 1: O-Brien Flemming boundary targets along with sample size provided the allocation ratio is maintained as "one" throughout the study.

Randomisation – Adaptive randomisation

This study involves a group sequential response adaptive randomisation procedure in which every participant will have an equal chance to be allocated to either of the study arms at the start of the study. The randomisation procedure used will be simple stratified randomisation based on the emetogenic risk of cancer chemotherapy regimen. The randomisation codes will be generated using R. We have four analyses (3 interim analyses and one final analysis) during the study as described in the sample size calculation section. Depending on each interim analysis result, the allocation ratio will be changed, and more patients will be allocated to the well-performing arm. If the z value of the effect size is less than one at an interim analysis, the trial will continue with a 1:1 allocation ratio. If the z value falls between 1 and 1.5, the allocation ratio will be changed to 1.5:1. If the z value between 1.5 to 2, the allocation ratio will be 2:1. If the z value falls beyond 2, the allocation ratio will be 2.5:1. The sample size will be altered if the allocation ratio is changed during an   interim analysis as follows.

Group with favourable response=(n(R+1))/2 

Group with unfavourable response=(n(R+1))/2R

n = Expected sample size per group if the allocation is equal between the groups.

R = Allocation ratio

 Blinding & Allocation concealment

This study is a double-blinded randomised clinical trial. Blinding will be maintained by using a similar-looking placebo. The assessor will be blinded regarding the groups of allocation of the participant. The assessor will not be involved in the analysis of data and interpretation of results. Allocation concealment will be done by allotting numbers to the randomisation chart generated. These numbers will be labelled on the drug containers.

Participant recruitment

Once the participant who is planned for moderately to highly emetogenic cancer chemotherapy consents for the study, an assessment of eligibility will be done by

• History and physical examination

• ECOG performance status

•  Liver & renal function tests and serum electrolytes

Administration of Treatment

After checking for the eligibility criteria, the baseline "nausea and vomiting" questionnaire is filled. Based on the allocation, either a starch-filled placebo capsule or a capsule with a combination of pregabalin 75mg with olanzapine 5mg along with the standard of care ondansetron 8 mg and dexamethasone 8 mg will be given to the participant one hour before the administration of chemotherapeutic agents. The participants will be provided four more capsules and instructed to take them daily (one capsule per day) for four days before going to bed.

Assessment of outcomes

After 24 hours, the patient will be contacted over the phone and enquired about the nausea severity (VAS score) and vomiting episodes. This process will be repeated for four more days. At the end of the 5th day, the patient will be asked to visit the Department of Pharmacology to evaluate the compliance and administer the FLIE questionnaire to assess the effect of the intervention on patient quality of life.

Assessment of Adverse event

The participant will be asked for the occurrence of any adverse event with a nonleading question initially.It will be followed with leading questions to assess the specific adverse events pertaining to the experimental drugs as follows.

• Undesired sedation and increased appetite will be assessed daily in terms of VAS scoring.

• Any disturbance in vision.

Statistical analysis

The statistical analysis will be done using R 4.1. The continuous data like VAS score, age, etc., will be expressed as mean (SD). The categorical variables like presence or absence of nausea/vomiting, sex, etc., will be represented as proportions. The data will be analysed using both the frequentist and Bayesian approaches. For the frequentist method, the response rate comparison is done with the z score for proportion providing the z values in each interim analysis. These z values from the intention-to-treat analysis will be used to assess boundary breach in O-Brien analysis and determine the allocation ratio for the forthcoming interim analysis. The continuous outcomes like change in VAS score and the score from the questionnaires will be assessed using an unpaired t-test. Continuous paired outcomes will be evaluated using paired t-test. The VAS scores for nausea and vomiting over five days will be assessed using mixed-effect models for longitudinal data. Both intention-to-treat analysis and per-protocol analysis will be performed along with the sensitivity analysis for the primary outcome. Replacement of the missing values will be dealt with using multiple imputation techniques or an expectation-maximisation algorithm. Adverse events analysis will be performed using chi-square analysis. Predictive probability and credibility interval for each group will be obtained using Bayesian analysis. Subgroup analysis will be performed to evaluate the efficacy of drug combination in terms of sex, age, previous history of nausea and vomiting and history of alcohol consumption.

4.Rationale of the Study

Olanzapine at a dose of 10 mg can cause unwanted sedation. A recent studies suggested that the dose can be reduced to 5mg without reducing efficacy . The standard treatment guidelines recommend using NK 1 receptor antagonist, but it has many drug-drug interactions as they have the property of inhibiting CYP-3A4 and CYP2C9 enzymes in a dose- dependent manner. Moreover, their cost burden is a significant concern inhibiting their use in patients with low socioeconomic status. So the search for an alternative treatment regimen that is cheaper and less CYP interaction is needed. Despite the newer antiemetic regimens, the control of nausea is not optimal. We theoretically consider that there might be a synergism as one drug (pregabalin) will reduce the release of neurotransmitters, and another drug (olanzapine) will block the action of the released neurotransmitters. So in this study, we will check the role of pregabalin (75 mg) combined with the olanzapine 5mg in the prevention of Chemotherapy Induced Nausea and vomiting especially Nausea control.