Opt-IVF Multi-center Trial 2 Including All Protocols

Aim: A Multi-center Randomized Clinical trial to determine the effectiveness of using the proposed decision support tool for ALL PROTOCOLS for each patient's customized optimal drug dosage profile.

Subject Enrollment

There are 13 centers involved in this trial. The recruitment plan designed is study-specific and site-specific. The investigators plan to recruit patients by reaching into the each center's own database of patients. Each center carries out on average 800 IVF cycles per year. PIs anticipate easier recruitment since they are not excluding any IVF patients, whether they are low, high, or average responders undergoing any protocol or patients with PCOS or not. The in-reach activities to be used include:

Chart review from site patient database Establishing an internal referral network with all physicians within the site Educational meetings with providers and physicians Educational sessions with patients and relatives Financial assistance to patients undergoing the intervention Training staff to facilitate ICF process by o Effectively conveying risks and benefits Creating an open environment Monthly calls to site physicians and staff to convey success stories

The retention activities include:

Planning appointments with sites for initial/follow-up visits and providing flexible/convenient appointment times Spending extra time with patients to understand their concerns and address them effectively Providing a summary of laboratory results to patients to demonstrate progress Appointment reminder services Travel assistance for patients/family members/caregivers, and patient compensation services Identifying the warning signals for potential drop-outs and working closely with these patients to support them to continue on the study Study Design and Procedures

This is a multi-site cohort with two arms (one arm for the intervention and one arm for the traditional approach) clinical trial involving 60 patients per center i.e. 780 patients total.

This task includes thrteen Indira IVF centers all over India, and all the analysis will be carried out at the Stochastic Research site. Dr. Urmila Diwekar will be an investigator from the Stochastic Research Technologies LLC, and Dr. Bhoi will be an investigator from the Indira IVF. Apart from these two investigators, there will be a post-doctoral fellow from Stochastic Research Technologies LLC and medical personnel from the Indira IVF centers will be involved in this proposed clinical trial.

In this study, Opt-IVF will be using the patient's age and day three serum day AMH and AFc levels to decide the starting dose for the patient's cycle. Opt-IVF will use the first day and day 5 data collected (Follicular size distribution, estrogen levels) for that patient to determine the optimal dosage profile for the entire cycle for that patient with the help of the decision support tool OPTIVF for this intervention in the clinical trial.

Primary and secondary outcomes: The study's primary outcome measures will be: the proportion of women with an appropriate number of retrieved oocytes), total hormonal dosage employed during the cycle, and serum oestradiol concentrations on rHCG day, no.of ultrasounds, no. of M2s, no. of embryos obtained, no. of grade A embryos and pregnancy rate.

The secondary outcome measures considered will be: the number of growing follicles (≥11 mm) on rHCG day; the number of large ovarian follicles ≥17 mm on rHCG day embryos transferred, fertilization rate, implantation rate, clinical pregnancy rate, OHSS rate, the number of cryopreserved embryos and the proportion of patients with cryopreserved embryos.

Data sources

Data Sources will be View Point (GE) and Syngo Dynamics (Phillips) for the sonographic images and Cerner for medical record information.

Expected Risks/Benefits

Current practice is to use every ultrasound and blood test every day to determine the dosage for that day that patient by the attending physician. The proposed decision support tool will be able to predict the patient's response to the hormonal dosage given each day of an IVF cycle using data from the first two days of that particular cycle. It will also provide a personalized optimal dosage profile for each patient, maximizing the outcomes and reducing medications, testing, and side effects. Based on earlier preliminary results, the tool use is expected to reduce the daily testing up to 75%, reduce the dosage up to 40%, thereby reducing the side effects like overstimulation, and is expected to provide equal or better outcomes for the patients. PI's group have carried out retrospective studies with 150 patients in the USA and India. The investigators also carried out two prospective clinical trials in India. Both studies support the above conclusion about the methodology. In this study, the investigators will be conducting prospective clinical studies for this product in the United States. This will establish the efficacy of this tool for use in clinical practice. There is no additional risk with the use of this decision support tool other than the possibility of an unsuccessful cycle, a risk that exists with current practice also. The same medications are used at dosages 20-40% lower, thus reducing the side effects of the medications.

Data Collection and Management Procedures

Collecting the de-identified patient data without any PHI. All the information will be digitized and managed securely (as needed), so it can be analyzed and researched. Proprietary file formats may be used in this process.

Data Protection Throughout the research work of this project, a large volume of data at various levels of detail. All research members will adhere to the highest standards for protecting subjects' privacy and the confidentiality of any information collected. Only properly trained and authorized individuals may access data with personal identifiers at each Indira IVF center. Nobody from stochastic research technologies will have access to personal identifiers.

Backups The company provide automated computer backups. Web Server It has been planned to disseminate and maintain project information through Stochastic webservers. These are all hosted on a Windows Server, a VMware virtual machine instance.

Data Analysis Data analysis will be performed by the co-PI using Excel and Matlab software.

Quality Control and Quality Assurance Data is recorded and coded in an electronic database. Categorical data will be compared using Chi-square or Fischer exact test and ANOVA for frequency table of baseline characteristics. Odds ratios and 95% confidence intervals will be calculated, as well as 2-sided p values. A value less than 0.05 will be chosen for significance without adjustment for multiple comparisons. A biostatistician will be consulted for any statistical analysis beyond the scope of the PI or co-investigators.

1 Data and Safety Monitoring De-identified data collected is in an excel spreadsheet in a password-protected computer.

Statistical Considerations Sample size: Earlier clinical trial shows that for obtaining statistically significant outcomes for comparing dosage and grade A embryos a small sample 22 patients in each arm was sufficient. However, for other outcomes like number of M2s higher samples size may be needed. Based on earlier results, the number of samples size is calculated to be 60 patients in each arm. Therefore, a sample size of 390 in each arm will be used in this trial.

Analysis Methods: Between the groups, differences in continuous variables will be assessed with parametric or non-parametric statistics, as appropriate. Z-tests will be conducted comparing the proportion of individuals who reported an optimal outcome response.

Regulatory Requirements Informed Consent The OPTIVF trial will be registered with Clinical Trials.Gov as per ACTG guidelines. PIs have established IRB-approved consent forms. This consent form is based on regular or long-form. The form presents detailed information about the treatments, risks, and benefits, mitigation of risks, etc., in summary, complete details on all elements of consent. Dr. Bhoi is responsible for this.

Subject Confidentiality The confidentiality of participants will be safeguarded by the use of a de-identified dataset. A master data spreadsheet including any necessary PHI will be kept in a locked file on a secure drive. All data analysis will be performed on the de-identified dataset. The key for the identification will be kept in a third locked file separate from the data. Only the co-PI will have access to these files. Data will be entered into SPSS or other statistical software programs for analysis without subject identifiers.

Unanticipated Problems Any unanticipated problems will be reported to the IRB as soon as they are discovered.