Design of the study

The METRO study is a multicenter, phase 2, randomized, double-blind, superiority trial that compares mesoglycan 50 mg BID with a matching placebo. Using a placebo as a comparator is justified by the absence of scientific evidence on the efficacy and safety of any treatment in this setting. The METRO study is conducted in 19 Italian Units of Vascular Medicine, all with significant experience in the management of VTE. The Coordinating Centre will be the General Medicine Unit and the Thrombotic and Haemorrhagic Disorders Unit at the Department of Medicine—Padua University Hospital, Italy. The study protocol has been reviewed and approved by the Ethics Committee of the Padua University Hospital, Italy, and by those of each participating center.

Before inclusion, patients/participants are requested to sign a written informed consent form.

An independent data safety monitoring board (DSMB), blinded to the patient's treatment allocation, formed by three highly qualified clinicians in the field of VTE, will adjudicate all efficacy events, monitor patients' safety, and vouch for the data quality.

Study population and randomization

Subjects of both sexes, > 18 years of age, with objectively documented SVT of the lower limbs, initially extended for at least 5 cm, who had completed the standard 45-day course with fondaparinux 2.5 mg OD, were asked to participate in the study. Patients could only be included if they consented to participate within two weeks from treatment withdrawal. The complete list of inclusion and exclusion criteria is reported in Table 1.

Table 1 Inclusion and exclusion criteria for METRO study

After signing the informed consent, eligible subjects will complete the screening visit. Only patients who meet the inclusion and exclusion criteria will be included.

The included patient will first undergo bilateral color-coded Doppler ultrasound (CCDU) of the lower limbs to exclude concomitant involvement of the deep venous system or an extension of the index SVT. Furthermore, blood sampling for alanine-aminotransferase, aspartate-aminotransferase, creatinine, complete blood count, prothrombin time, activated partial thromboplastin time, and C-reactive protein will be performed.

A computer-based centralized randomization, with 1:1 allocation between mesoglycan and placebo, and random variable size blocks (4 to 6 units), stratified for age (< 60 y, >  = 60 y), and gender (F/M), will be run through a dedicated application (REDCAp). Patient recruitment is expected to be completed in 24 months.

Treatment

Patients will receive a hard capsule of 50 mg mesoglycan or a matching placebo twice daily for 12 months. Treatment packages will be labeled with a bare numerical code without reference to the content. The match between the numerical code and the package content will only be known to the packaging manager. Patients, investigators, the steering committee, and DMSC will be blinded to treatment group allocation. The blinding seal will only be opened after the database lock to allow the analysis of the study results, except for in the case of a medical emergency, for which it is considered essential to know the type of treatment.

Each participating center will keep a printed drug accountability form for each patient included in the study (drug amount delivered, consumed or returned).

Patient flow

Medical control visits will be scheduled 3, 6, 9, and 12 months after randomization for every included patient, during which a complete evaluation of their clinical status, compliance with treatment, concomitant treatments and adverse events will be performed. At each visit, patients will also receive the revised Venous Clinical Severity Score (rVCSS), which accounts for changes in disease severity over time and in response to treatment, and the VEINES/Sym-QoL questionnaire, a disease-specific quality of life instrument for chronic venous disorders of the leg [34,35,36,37,38,39]. A CCDU examination and a blood sample will also be scheduled for the 12 month visit.

At the end of the 12-month treatment cycle, patients will enter another 12-month clinical follow-up period, during which they should not receive any experimental treatment; for a total study time of 24 months. A final visit will be scheduled 24 months after randomization in patients who complete the study course.

Patients who meet exclusion criteria at any time during the study, become pregnant, do not comply with study purposes or withdraw their consent will be censored. Patients who withdraw experimental treatment will continue to observe according to the study protocol until the end of the study. The planned flow of the patient is shown in Fig. 1.

Fig. 1

Flow of the patients through the study

Study outcomes

The primary efficacy outcome of the study will be an objectively documented symptomatic recurrence or extension of SVT, symptomatic DVT (proximal or distal), asymptomatic proximal DVT, symptomatic pulmonary embolism (fatal / nonfatal), or death, whichever comes first, during the treatment period.

The secondary efficacy outcome will be: the occurrence of asymptomatic recurrence, extension, or asymptomatic distal DVT; the rVCSS score and the VEINES / Sym-QoL questionnaire; the impact of elastic compression therapy on SVT treatment; the association between CRP levels and primary efficacy outcomes; the natural history of SVT in patients allocated to placebo; the evaluation of the prolonged protective effect of mesoglycan after the first 12 months of treatment.

The safety outcome will be the incidence of major bleeding, as defined by the International Society on Thrombosis and Hemostasis (ISTH) criteria: fatal and/or symptomatic bleeding in a critical area or organ (e.g., intracranial, intraspinal, intraocular, retroperitoneal, intraarticular, pericardial, intramuscular with the associated compartmental syndrome); and/or in the case of clinically overt bleeding associated with a hemoglobin decrease of at least 20 g/L or requiring transfusion of 2 or more units of packed red blood cells [40].

Outcome assessment

Patients will receive detailed instructions on how to contact their referral center if symptoms suggestive of efficacy or safety events occur during the course of the study (treatment period and clinical follow-up).

Patients with suspected recurrent SVT or DVT will undergo CCDU. As previously described, whole-leg CCDU will be performed according to a standardized protocol [41]. Briefly, the proximal venous system of the lower limbs (common, superficial, and deep femoral veins, and popliteal vein to the trifurcation) is examined first with the patient lying supine; then the distal venous network (tibio-peroneal and muscular veins) is scanned, with the patient sitting. We do not routinely investigate the anterior tibial veins. Therefore, vein incompressibility is the only diagnostic criterion used to rule in proximal and/or distal DVT and SVT. Patients with symptoms suggestive of PE are investigated with Computed Tomography Pulmonary Angiography (CTPA) or Pulmonary Ventilation (V)and Perfusion (Q) scan. PE is adjudicated in the presence of an intraluminal filling defect involving the principal, lobar, or segmental branches of the pulmonary artery. If the patient dies, the cause of death is adjudicated as "likely" or "unlikely" due to PE based on autopsy findings, if available, or on clinical grounds.

Data Collection

Data will be recorded on an “Electronic Data Capture” (EDC) system, based on the “Research Electronic Data Capture” online platform (REDCap, produced and distributed by Vanderbilt University and “REDCap Consortium”), in compliance with article n. 13 of the GDPR (EU 2016/769) [42].

The personal data of the subjects enrolled will be treated with maximum confidentiality, according to the terms of Italian law (DL 211/2003 and subsequent amendments and additions) and GDPR [Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016].

In particular, the subject's personal data will be known only to the principal investigator and his collaborators who follow the patient and collect the treatment consent. Each enrolled subject will be distinguished only by a unique numeric identifier. The use and maintenance of the REDCap platform are guaranteed by an administrator who manages the user's privileges with a flexible and granular authorization system. REDCap applies the permissions granted to each user who connects via a web browser and the SSL encryption protocol. REDCap allows a complete "audit" of the procedures performed by each user by logging all operations on the data, including viewing and exporting. The operational control log stores the date, time, and operating behavior of the user, allowing a complete review and remote monitoring of the clinical study if necessary. All users accessing the EDC platform (the investigators, the members of their staff, and the staff of the Promoter) must complete a training event to increase the reliability, quality, and integrity of the data recorded in the EDC platform. REDCap implementations allow compliance with the most common industry standards and EMA requirements (GCP, Privacy-IT: DL 211/2003 and subsequent amendments and additions, GDPR, and FDA (21-CFR2-Part 11).

Statistical analysis and sample size

From the available data in the literature we hypothesize a risk of the occurrence of the primary efficacy outcome in the placebo group at the end of the treatment of 15% (that is, a cumulative proportion of 85% of subjects without events) and a 50% decrease in risk in the mesoglycan group (HR = 0.5) [3, 5, 9].

A two-sided log-rank test with an overall sample size of 650 subjects (of which 325 are in the placebo reference group and 325 in the mesoglycan treatment group) achieves a power of 90%, at a significance level (alpha) of 0.025, to identify a difference of 7.2%, between 15.0 and 7.8% (the proportions of subjects with primary endpoint in the placebo and mesoglycan group, respectively) after 12 months of treatment. This corresponds to a hazard ratio (HR) = 0.50. Subjects will be enrolled in the study for 24 months, and 50% of enrollments will be completed in month 12. The calculation considers the loss of subjects enrolled during follow-up of 10.0% in the placebo group and 10.0% in the mesoglycan group. 650 subjects will be enrolled and randomized 1:1 in the two placebo and mesoglycan arms.

This number seems large enough to guarantee the study an adequate “power” (> 0.80) even if the risk of observing the primary endpoint in the placebo group (hazard rate control) is the difference in efficacy between the two treatments (hazard ratio) that will deviate significantly from those hypothesized.

However, an “interim-analysis” is prepared when 3/4 of the subjects (n = 488) are enrolled so that the “Data and Safety Monitoring Committee” can reformulate the number of subjects to be enrolled (“resampling”) to reach the preset study power (90%) if the number of primary endpoint events observed up to then deviates significantly from the expected for the two groups, i.e., if the HR differs from the estimated one (50%) exceeded 5% in absolute value (a lower deviation seems tolerable based on the above simulation).

The alpha error is, therefore, conservatively set at 0.025 (O'Brien-Fleming method).

For the interim analysis, a tabular description of the frequency of the primary efficacy endpoint will be provided to the Committee, stratified by treatment group but without its label (i.e., without identifying the nature of the treatment assigned to each group).

Statistical considerations

All variables will be descriptively analyzed with the appropriate statistical methods: for categorical variables using frequency tables and for continuous variables with sample statistics (e.g., mean, median, standard deviation, minimum and maximum value, 25° and 75° quartile).

Unless otherwise specified, all statistical tests will be 2-tailed and at a 5% significance level.

All the primary and secondary efficacy and safety analyses will be performed in the modified intention-to-treat (ITT) population, including all randomized subjects who received at least one dose of the study drug.

Further supporting analysis will be performed in the per-protocol population, which will include all randomized subjects who were complying with treatment, that is, taking at least 65% of the intended dose units of the study drug.

Subjects who are randomized and who have taken any amount of study drug will not be replaced for any reason. Subjects who will be randomized but withdraw their consent before taking the study drug may be replaced, although they will not lose their unique identifier.

Subjects who leave the observation without manifesting the endpoint will be considered censored, participating in the risk during the observation period. For subjects who do not appear for clinical evaluations at 12 and 24 months of observation, every effort will be made to know a possible outcome of interest for the study, making them evaluable for endpoints.

Treatment compliance

Compliance with oral treatment can be used as a stratification variable (< 65% and >  = 65%) and defined as compliance = the total number of tablets actually taken / the total number of tablets expected*100.

Final analysis of the primary endpoints

The primary cumulative efficacy endpoint will be the occurrence of the first event among the event types considered. The hypothesis of superiority of the primary cumulative efficacy endpoint will be tested by multivariate Cox proportional hazards regression using as covariates the assigned drug, age at randomization, sex, compliance with treatment, and the additional risk factors for DVT and DVT/PE found to be significantly associated in the univariate analysis (between obesity, varicose veins, cancer, standing or prolonged immobilization, recent surgery, severe infections, prior DVT/PE, thrombophilia, chronic heart failure NYHA III-IV, use of estrogen-progestogens). All variables participating in the model will be tested for the assumption of proportionality of the risks with the usual graphical method. In addition, a significant subset of variables will be chosen with the Wald method and in a “forward stepwise.”

The crude and adjusted cumulative proportion of event-free subjects over follow-up in the two treatment groups will be described with Kaplan–Meier plots.

Final analysis of secondary endpoints

For the analysis of the secondary efficacy endpoints, the following will be used:

1.

the Kaplan–Meier plot of the cumulative occurrence over time and the log-rank test to compare the two treatment arms,

2.

for the rVCSS scores, analysis of variance for repeated measures will be used (all available assessments for within-subject effects and the 2 treatment groups as between-subjects effect).

3.

in the case of the VEINES/Sym-Qol scores, the analysis of variance for repeated measures will be used (all evaluations available for the within-subject effect and the 2 treatment groups as the between-subjects effect).

4.

The impact of elastic compression therapy in the various endpoints will be analyzed by inserting it as a stratification factor in the related analyzes.

5.

The predictive contribution of CRP levels (subdivided by tertiles) to the primary efficacy endpoint will be analyzed by insertion as a stratification factor in relative analyzes.

Final analysis of the security endpoints

Analysis of the primary safety endpoint will be performed with the continuity adjusted chi-square test.

Standard summary of safety data related to adverse events and significant changes in laboratory data will also be provided.