Harmonized clinical trial methodologies for localized cutaneous leishmaniasis and potential for extensive network with capacities for clinical evaluation

Introduction Progress with the treatment of cutaneous leishmaniasis (CL) has been hampered by inconsistent methodologies used to assess treatment effects. A sizable number of trials conducted over the years has generated only weak evidence backing current treatment recommendations, as shown by systematic reviews on old-world and new-world CL (OWCL and NWCL). Materials and methods Using a previously published guidance paper on CL treatment trial methodology as the reference, consensus was sought on key parameters including core eligibility and outcome measures, among OWCL (7 countries, 10 trial sites) and NWCL (7 countries, 11 trial sites) during two separate meetings. Results Findings and level of consensus within and between OWCL and NWCL sites are presented and discussed. In addition, CL trial site characteristics and capacities are summarized. Conclusions The consensus reached allows standardization of future clinical research across OWCL and NWCL sites. We encourage CL researchers to adopt and adapt as required the proposed parameters and outcomes in their future trials and provide feedback on their experience. The expertise afforded between the two sets of clinical sites provides the basis for a powerful consortium with potential for extensive, standardized assessment of interventions for CL and faster approval of candidate treatments.


Introduction
Progress with the treatment of cutaneous leishmaniasis (CL) has been hampered by inconsistent methodologies used to assess treatment effects. A sizable number of trials PLOS Neglected Tropical Diseases | https://doi.org/10.1371/journal.pntd.0006141 January 12, 2018 1 / 12 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 conducted over the years has generated only weak evidence backing current treatment recommendations, as shown by systematic reviews on old-world and new-world CL (OWCL and NWCL).

Materials and methods
Using a previously published guidance paper on CL treatment trial methodology as the reference, consensus was sought on key parameters including core eligibility and outcome measures, among OWCL (7 countries, 10 trial sites) and NWCL (7 countries, 11 trial sites) during two separate meetings.

Results
Findings and level of consensus within and between OWCL and NWCL sites are presented and discussed. In addition, CL trial site characteristics and capacities are summarized.

Conclusions
The consensus reached allows standardization of future clinical research across OWCL and NWCL sites. We encourage CL researchers to adopt and adapt as required the proposed parameters and outcomes in their future trials and provide feedback on their experience. The expertise afforded between the two sets of clinical sites provides the basis for a powerful consortium with potential for extensive, standardized assessment of interventions for CL and faster approval of candidate treatments.

Author summary
The term 'cutaneous leishmaniasis' (CL) includes a range of manifestations affecting the skin caused by Leishmania parasites across several continents. While not life-threatening, CL can be invalidating and disfiguring, or become complicated. Today, there is no satisfactory treatment for CL that is effective and safe. Faced with no investments into developing drugs for CL, clinical researchers have tried many treatments over the years, but little progress has been made. One of the reasons is the lack of standardized methodologies in conducting these trials which makes it difficult to collate and compare results. Clinical researchers now realize that their efforts can be brought to fruition if common methodologies are available and applied. This paper summarizes the principles and parameters agreed upon by researchers of how to identify patients and how to measure treatment effects in a way that will make it possible to gather convincing evidence of whether a treatment works or not. Adhering to these principles will allow faster progress towards offering better care to patients with this neglected disease.  [1,2]. In 2014, over 153 000 cases were reported to WHO from 10 high-burden countries [3,4]. Progress with the treatment of CL has been hampered by lack of investments in drug discovery and development, but also by the inconsistent methodologies that have been used to assess treatment effects [5]. This has resulted in significant scientific and financial waste, as a sizable number of trials conducted over the years have generated only weak evidence for treatment recommendations.

Introduction
These weaknesses were exposed by two Cochrane systematic reviews on Old-World [6] and New-World [7] CL (OWCL and NWCL; the latter recently updated [8]). To correct these shortcomings, a series of steps were set in place towards achieving consensus on the main parameters that would help establish standardized, generally adoptable criteria in clinical investigations. This process started with a consultation jointly organized by the Special Programme for Research and Training in Tropical Diseases (WHO/TDR) and the World Health Organization Programme for Leishmaniasis at the Neglected Tropical Diseases Department (WHO/NTD) held in 2009, which led to a guidance paper in 2013 [9] that aimed to (i) provide clinical investigators with guidance for the design, conduct, analysis and report of clinical trials of treatments for CL, whilst recognizing the complexity of the disease; and (ii) enhance the capacity for high-quality trials that fulfil the requirements of International Conference on Harmonization (ICH) and Good Clinical Practice (GCP) standards.
A network of clinical trial sites for NWCL (RedeLeish [10]) was started with support by the Drugs for Neglected Diseases initiative (DNDi) in 2014, and discussions are underway to extend this network to OWCL (jointly with TDR).

Methods
Basic parameters from the above-mentioned guidance paper were submitted to a group of OWCL and NWCL clinical trialists and discussed at workshops that took place in Tunisia (February 2016, hosted by the TDR regional training centre at Institute Pasteur, Tunis and organized by TDR) and in Brazil (June 2016, organized by DNDi). The meetings were attended by expert CL trialists representing 10 clinical study sites from 7 OWCL countries (Afghanistan, Burkina Faso, Ethiopia, Iran (Islamic Republic of), Morocco, Tunisia and Turkey) and 11 clinical study sites from 7 NWCL countries (Bolivia, Brazil, Colombia, Guatemala, Panama, Peru and Venezuela).

Consensus on key methodological issues in clinical trials of CL treatments
The degree of consensus and main issues are summarized in Table 1 for OWCL and NWCL, along with the revised parameters after the two above mentioned consultations. A diagrammatic representation can be found in Fig 1. As for eligibility criteria, it was agreed that parasitological confirmation by visualization of the parasite (amastigotes in smears, promastigotes in culture) or molecular biology testing (primary PCR) is required for a patient to be enrolled in clinical trials; Leishmania species identification is not required for enrolment but is required for data analysis. The need for baseline safety tests (hematology, liver and renal function) depends on the risks associated with the treatment (route of administration and the phase of development), the drug's chemical class, and the perceived or known liabilities of the treatment (expected toxicity).
As for the efficacy parameters, there was consensus about re-epithelization for ulcerated lesions and flattening for non-ulcerated lesions as primary efficacy measures. The majority of participants was for adding absence of induration as an efficacy parameter (though more difficult to standardize), while redness (inflammation) was thought to be not sufficiently reliable.
Even though the natural history and treatment response vary across the range of old and new world Leishmania species, it was agreed that initial cure should be assessed at Day 90-100 (Day 0 being the day of enrolment and Day 1 being the first day of treatment) since it provides the best chances to assess success or failure. In OWCL treatment trials, an additional earlier assessment at Day 42 should also be conducted and reported to capture the earlier clinical response observed in these species.
In addition, OWCL participants identified the need to document more clearly and quantify the rate of self-healing in L. major, in order to better inform decisions on follow-up and study design, such as the assessment of time-to-heal particularly after topical treatment, as a secondary outcome-which would require multiple assessments. NWCL participants discussed the need to collect evidence towards a future definition of "early failure" (before Day 42) based on the type of intervention/treatment being evaluated. While it was acknowledged that, in some instances, treatment is provided until the lesion is considered cured (especially when evaluating topical or intralesional treatments), efforts should be made to report the number of cured subjects at day 42 and the initial cure at day 90.
Final cure should be assessed at Day 180 (6 months after initiating treatment); a 12-month follow-up was not deemed necessary. Nevertheless, NWCL participants identified the need to assess the ideal time of follow-up for final cure, and document the rate of late-responses and relapses between days 90-180 (3-6 months). This would provide important elements to understand the cost-effectiveness of a 6-month follow-up, and inform study design.
There was almost general agreement that time of follow-up is counted starting from the first day of treatment and not from the end of treatment. The main issue was how to deal with treatments of different duration. For instance, systemic antimonials are given for 14-30 days (see Tables 2 and 3); an initial assessment at day 42 counting from treatment start means 12-15 to 28-31 days after the end of treatment, compared to e.g. thermotherapy, which may be given in one single treatment.
It was also agreed that treatment failure should be defined at Day 42 as less than 50% re-epithelialization (if an ulcer) or flattening (if a non-ulcerated lesion); and at Day 90 as less than 100% re-epithelialization or flattening, respectively.    It was suggested that the presence and grading of scars (cosmetic effects) should be assessed in a standardized way and included in future long-term treatment evaluation, as this represents an important parameter for patients because of the related stigma and social consequences. The results of qualitative studies using in-depth semi-structured CL patient interviews aimed at understanding the CL patient's needs and expectations from treatment (paper in preparation) will also help inform study design.

CL trial site characteristics and capacities
Site characteristics for OWCL and NWCL are summarized in Tables 2 and 3 respectively.
About the sites. 'Site' here refers to a single or multiple treatment sites with different catchment areas covered by one group. Of the OWCL sites, 3 are referral centers, 2 primary and 5 both; for NWCL, 7 are referral and two both primary and referral. Capacity for good clinical practice (GCP) trials exists in 8/10 OWCL sites and 9/9 NWCL sites.
About the disease: OWCL. Species and diagnosis: Cases of both L. major and L. tropica cases are seen at 6/10 sites (one also L. infantum), while 2 sites have one species each and 2 have L. aethiopica. Parasitological diagnosis by direct smear is done at all sites. Capacity for polymerase-chain reaction (PCR) exists at 8 sites and culture at 5, though techniques are not always used routinely.
Age and Sex: All ages are affected in 7/10 sites. L. aethiopica affects mostly older children, adolescents and young adults. OWCL affects equally women and men, with more men being seen only in Iran.
Endemicity: Six out of 10 sites have stable transmission, while 4 see patients from both emerging and stable foci. There is no obvious pattern relating age and transmission. Burkina-Faso has a stable focus in Ougadougou and a newer one in Bobo-Dioulasso.
Setting: Patients seen at 6/10 sites are from both rural and periurban settings. In Burkina-Faso the Ougadougou focus is periurban, while the one in Bobo-Dioulasso is rural.

Discussion
The consensus reached among participants during the two meetings allows standardization of future clinical research across OWCL and NWCL sites-a major issue which has hampered our collective ability to generate strong evidence for treatment guidelines and policy. We encourage CL researchers to adopt and adapt if so required the proposed parameters and outcomes in their future trials. Furthermore, the expertise afforded between the two sets of clinical sites provides the basis for a powerful consortium with potential for extensive, standardized assessment of interventions for CL.