This was a Phase III, multicentre, randomized, investigator-masked, cross-over, comparative, non-inferiority trial (clinical trial identification number: 848,300,144/0103/1 - POP03; IND number/EudraCT number: 2018-001727-39). The study was conducted between January 2019 and March 2020 in five sites in two countries: three sites in Hungary and two in Russia. The sites were evaluated, selected, and submitted for approval to the ethics committees and competent authorities. The study was designed to assess the non-inferiority of latanoprost Polpharma (test), a generic, preservative-free latanoprost 0.05 mg/ml solution, versus Xalatan® (reference), a latanoprost 0.005% ophthalmic solution, in accordance with the “Guideline on the Choice of the Non-Inferiority Margin”, Sect. 3.2. “Two arm trials: test and reference” EMEA/CPMP/EWP/2158/99 (CHMP, 2005). Our study adheres to CONSORT guidelines.

Latanoprost s delivered in multi dose container: a high-density polyethylene bottle with 3 K dropper pump. The 3 K pump system is a patented dispensing apparatus with the main purpose to protect the contents from microbiological contamination. The fluid path is protected from microbiological contamination by a silver coil in the 3 K pump tip.

The study was investigator-masked, but since the primary packaging of the reference product was different from that of the test product, each site assigned unmasked personnel who were in charge of handling, distribution, and return of the investigational products. Both test and reference products were packaged and labelled identically with the exception of primary packaging. The patients were also instructed not to reveal the identity of the investigational products that they have been assigned to the investigator or other personnel involved in the evaluation of the study outcomes.

This trial had a cross-over design. The sites received medication boxes, with printed randomization numbers. The designated site staff assigned the lowest available number to the patient and handed out the corresponding medication box to the patient. The patients who met the eligibility criteria on screening day were randomly assigned in a 1:1 ratio (using block sizes of 4) to be treated for 29 ± 1 days with one of the two investigational products (period I). After a washout period of at least 28 days, the same patient was treated for 29 ± 1 days with the other investigational product (period II) (Fig. 1). The wash-out phase of 4 weeks was chosen based on the FDA draft guidance on Brinzolamide [4].

A placebo group was not included, since there is ample evidence about the placebo effect in studies addressing the effect of topical medication on IOP using the Goldmann Applanation Tonometer. The effect is only a fraction of the effects observed in the study [5].

Cross-over design and treatment periods. Test product Latanoprost Polpharma; reference product Xalatan®

The primary endpoint was to evaluate the non-inferiority of the test product when compared to the reference product with respect to the differences in the mean diurnal IOP in the study eye on Day 1 (baseline) and Day 29. The non-inferiority margin was set to 1.5 mmHg for the difference in treatment effect between the test and reference product, which has been widely used in comparable studies with IOP lowering agents.

The primary efficacy parameter was the mean diurnal IOP. IOP was determined using a Goldmann Applanation Tonometer under local anaesthesia and following established procedures at the site. Two IOP measurements were done at each time point and the mean thereof was recorded as the IOP value at that corresponding time point. Since the IOP is known to vary naturally over the course of the day, 4 different IOP measurements were taken (approx. at 08.00, 12.00, 16.00, and 20.00 h) at baseline and at the end of the treatment in each treatment period to calculate the mean diurnal IOP. These four timepoints were chosen based on the data by Camras, who showed that a plateau of low IOP was reached after the third measurement at approximately 16.00 [2]. The IOP lowering effect was then calculated as the difference in the mean diurnal IOP in the study eye between Day 1 and Day 29.

The secondary endpoints included efficacy, ocular tolerance, safety, and usability. With respect to efficacy, we investigated the non-inferiority of the test product when compared to the reference product with respect to the differences in IOP at each measurement time point on Day 1 and on Day 29. We also investigated the difference between the investigational products with respect to ocular comfort level score and conjunctival hyperaemia on Day 1 and Day 29. The scale to rate the ocular comfort level was as follows: 0 = no discomfort; 1 = mild discomfort disappearing within 20 min after treatment; 2 = moderate discomfort, i.e., no medical intervention needed, expected to disappear within one hour; 3 = severe discomfort, medical intervention needed; 4 = very severe discomfort, medical intervention needed, use of investigational product is interrupted or discontinued. Ocular comfort was assessed immediately post dose (0 min), 5, 10, and 20 min post dose on Day 1 and Day 28 in both treatment periods. Hyperaemia was evaluated by performing a slit lamp examination 20 min post dose. It was described using the following scores: 0 = no conjunctival hyperaemia, vessels normal; 1 = sporadic vessels clearly injected above normal; 2 = diffuse red eye with individual vessels dilated but still discernible; 3 = intensive red eye with strong dilation of conjunctival vessels, which are no longer easily discernible.

To assess safety, we investigated the difference between the investigational products with respect to general safety as assessed by vital signs and the incidence and nature of adverse events (AEs). Vital signs were determined by measuring blood pressure and heart rate after 3 min in sitting position following established procedures at the site. AEs were coded according to MedDRA (English Version 23.0). The patients evaluated the usability of each of the delivery devices by means of a questionnaire with eight scoring questions (scale 1–10) and three free text questions, which the patients completed at the end of each period.

The study had the ethical and regulatory approval from the National Institute of Pharmacy and Nutrition (OGYEI, OGYEI/41,779- 11/2018) and the Ethics Committee for Clinical Pharmacology (KFEB) of Hungary and from the Ministry of Healthcare of the Russian Federation (MOH of Russia) prior to the beginning of the study (642/25.12.2018). The study was conducted in accordance with the ICH guidelines for Good Clinical Practice (GCP, E6) and the Declaration of Helsinki (Version 6 of 64th General Assembly of WMA in Fortaleza, Brazil 2013). Data management was conducted in compliance with Good Clinical Data Management Practices.

A total of 53 patients aged 18–75 years old with open-angle glaucoma or ocular hypertension were recruited. The inclusion criteria were as follows: age: 18–75 years old; provision of signed and dated Informed consent; general health conditions not interfering with participation in the study (e.g. blood pressure); female patients of childbearing age should either be using acceptable methods of birth control or be heterosexually inactive (abstinent) for at least 28 days prior to the first dose and throughout the study); ocular hypertension or POAG in both eyes: mean diurnal IOP measured at -12, -8, -4, 0 h pre-treatment on Day 1 must be higher than or equal to 22 mmHg, and lower than or equal to 34 mmHg (naïve or untreated, i.e., after washout); not on any ophthalmic pressure-lowering medication, or able to be withdrawn from current pressure-lowering medications for the washout periods; no clinically significant or progressive retinal disease as determined by dilated peripheral retinal examination done at screening; no concomitant use of any topical ophthalmic medication other than artificial tears; no ocular glucocorticoids in the previous 3 months; no ocular trauma, surgery, inflammation or infection, no corneal foreign body in the previous 3 months; no systemic medication that may alter IOP in the previous 30 days (e.g., beta blockers, calcium channel blockers, ACE inhibitors, prostaglandins, etc.) or expected to continue the current treatment with these medicinal products on a stable regimen for the duration of the study. Patients who were contact lens wearers had to agree not to use contact lenses for the duration of the study.

Patients were excluded from the study if they fulfilled any of the following criteria: a corrected visual acuity of less than distance Snellen 20/100 corresponding to decimal 0.20 or log MAR 0.70 in both eyes; evidence of acute ocular infection, corneal foreign body, or ocular inflammation within 3 months of the screening visit; a history or evidence of severe inflammatory eye disease in one or both eyes, especially conjunctival hyperaemia score at inclusion > 0; previous significant ocular trauma, laser or incisional surgery within 3 months of the screening visit; traumatic cataract surgery with an open posterior capsule or any patient with an anterior chamber intraocular lens implant or aphakia; IOP in either eye exceeding 34 mmHg (mean diurnal at Day 1: -12, -8, − 4, 0 h); IOP in either eye greater than 34 mmHg at Day 1 (mean diurnal IOP measured at -12, -8, − 4, 0 h pre-treatment); any corneal abnormalities preventing reliable applanation tonometry; central corneal thickness < 450 μm or > 600 μm; patients at risk of angle closure or evidence of acute, intermittent, or chronic angle closure; forms of glaucoma resulting from conditions other than primary open-angle glaucoma or ocular hypertension, such as pigmentary or pseudo-exfoliative glaucoma; pupil with inadequate ability to dilate sufficiently for peripheral retinal examination; history or evidence of Herpes simplex keratitis; patients with known risk factors for macular oedema; pregnant or nursing women or women who intend to become pregnant during the trial; patients who have participated in another research study for an investigational product or investigational medical device within 30 days of the screening visit; history of drug or alcohol abuse within the last 6 months; a history of hypersensitivity to latanoprost, or any component in the formulation of the products being tested; history of evidence of any medical condition that would, in the opinion of the investigator, make the patient unsuitable for the study (i.e. severe hepatic, cardiovascular or renal impairment); systemic medication that may alter IOP in the previous 30 days if the treatment regimen with these medicinal products is changed during the study.

Informed consentwas obtained from all the patients included in the study prior to any study-related activities and in accordance with all applicable regulatory requirements. The investigator and/or his/her designee orally informed every patient in addition to the written patient information about all aspects of the patient’s participation in the study. The competent ethics committees and regulatory authorities approved the written patient information and informed consent form, according to the local regulations of the European Union, Hungary, and Russia.

On Day 1, the patient reported to the clinic, where the IOP was measured at four different timepoints and the first dose of the assigned investigational product was administered. Thereafter, patients were instructed to instil one drop of the investigational product into the affected eye(s) once a day, with an interval as close as possible to 24 h between 20:00 and 22:00. The patients also were given diary cards which they were asked to complete at home in order to document medication compliance.

Follow-up visits took place on Day 14, Day 28, and Day 29 of both periods. Patient compliance was documented in the patients’ diaries and by weighing the bottles at the beginning of each treatment period and at the end of the treatment period by the unmasked site staff. In addition, patients were contacted by phone on Day 7(± 2) to enquire about their well-being and to assess compliance. A maximum of 20% missed doses during one treatment period was considered acceptable for per protocol evaluation.

Patients experiencing any of the following did not receive further dosing of test or reference product: confirmed pregnancy or wish to become pregnant; mean diurnal IOP higher than 34 mmHg in either eye; anaphylaxis; severe adverse reaction; severe inflammatory eye disease in one or both eyes; non-compliance to study procedures.

A sample size of 50 patients (including a 20% drop-out rate) was calculated to provide 90% power that the 95% confidence interval of the difference in change of mean diurnal IOP from baseline to Day 29 between the two products will be within 1.5 mmHg, assuming a treatment effect of 8 mmHg IOP reduction and a standard deviation of 3 mmHg and no real difference between the two products. Assuming an attrition rate of approximately 20% (drop-outs, protocol deviations) a total of 42 evaluable patients was found to be needed. The sample size was calculated using the ExpDesign Studio 5.0.2 Software referenced in the book by Chang [6].

A total of 49 patients were included in the analysis (safety [SAF] population). Of these, 47 were evaluable for the efficacy endpoints (intention to treat [ITT] population). The efficacy analysis included data from all patients completing the study according to the protocol without major protocol deviations (per protocol [PP] population), as well as from all patients completing the study (ITT population). The safety analysis included data from all patients that received at least one dose of the investigational products (Fig. 2).

The non-inferiority of the test product in comparison with the reference product was tested using a mixed linear model. The goal was to reject the null hypothesis H0 at the one-sided significance level α = 0.025. The approach to testing non-inferiority was to use the two-sided 95% confidence interval (CI) for the difference (D) of the effects of the test and reference product:

D = Effect (test) -Effect (reference) > -1.5 mm Hg.

This was done using the following non-inferiority hypotheses:

H0: Test – Reference ≤ – 1.5 mmHg (Test is inferior to Reference).

Ha: Test – Reference > – 1.5 mmHg (Test is not inferior to Reference).

The effect of treatment (primary efficacy parameter) was calculated as the difference between the mean diurnal IOP in the study eye after 29 ± 1 days of treatment and baseline (pre-treatment). As secondary efficacy parameters, the effect of treatment for each of the four measurement time points of the diurnal curve was calculated.

The efficacy analysis was performed on the PP population and repeated as sensitivity analysis on the ITT population to assess the robustness of the study results. Safety and ocular tolerance data were analysed descriptively.

The study procedures had to be adapted because of the COVID-19 pandemic, starting mid March 2020. Due to a specific request by the authorities in Hungary, the last monitoring visits and the close-out visits in this country had to be performed remotely, for which specific guidance documents were issued.

In modification of the protocol, the non-inferiority of the test product when compared to the reference product was investigated not only with respect to the differences in mean diurnal IOP but also with respect to the difference for each measurement time point at baseline (-12, -8, -4 and 0 h before treatment) and Day 29 (12, 16, 20 and 24 h after treatment the previous day) was determined. This modification was reflected in the statistical analysis plan. The change was introduced because it became apparent that authorities in Europe and overseas increasingly requested this data set for the non inferiority assessment of IOP lowering agents. The change has no impact on the validity of the study and its outcomes.