Unconventional therapies in ocular adnexal lymphomas

Keywords::

• antibiotic therapy
• chlamydia
• infection
• lymphoma

Ocular adnexal lymphomas (OALs) represent relatively rare lymphoproliferative disorders but have a rising incidence [1]. The most frequently occurring OAL histotype is represented by extranodal marginal zone B-cell lymphomas (OAMZLs), which are indolent tumors, usually presenting as stage I–II disease. An intriguing issue involves the relationship of OAMZL with bacterial infection, in particular with Chlamydia infection. Our group reported a strong association of Chlamydophila psittaci (Cp) infection in nearly 80% of Italian patients with OAMZL. Among these patients, 40% showed molecular evidence of Cp infection in peripheral blood mononuclear cells (PBMCs) [2]. Cp is carried by monocytes/macrophages present in the lymphomatous tissue, as proven by immunohistochemistry, immunofluorescence, electron microscopy and laser capture-assisted PCR followed by direct sequencing [3]. Cp is living and infective both in conjunctival tissue and PBMCs [4]. In vitro isolation of this microorganism from patients with OAMZL fulfills Koch’s second postulate, enhancing the specific relationship between Cp and lymphomagenesis in this anatomical region.

A critical topic in OAMZL is the lack of consensus about the best upfront treatment for these disorders. Ideally, the optimal treatment should not be negatively counterbalanced, if not overcome, by side effects. In fact, the range of therapeutic options is heterogeneous, including at one end a ‘watch and wait’ policy and, at the opposite end, systemic chemotherapy, usually represented by a rituximab–cyclophosphamide–adriamycin–vincristine–prednisone (R-CHOP) regimen. A large proportion of patients are usually treated as a first-line approach with radiotherapy, with obvious implications related to the long-term complications of this strategy. The most recent intriguing therapeutic approach in OAMZLs is represented by the use of doxycycline in Cp-related lymphomas. Within one pilot study, an overall response rate of 48% has been obtained [5]. According to these results, doxycycline appears a cheap, fast and active therapy in Cp-related OAMZLs. On the basis of this premise, a multicenter Phase II prospective trial has been launched under the auspices of the International Extranodal Lymphoma Study Group (IELSG #27 trial) [6]. This trial will be completed in the forthcoming months, and the actual response rate to this antibiotic will soon be available. In the meantime, we strongly believe that the present use of antibiotics in OAMZLs still needs to be regarded as an experimental therapy, and therefore this approach should be discouraged outside clinical trials [7].

Many unanswered questions involve the issue of Cp-related OAMZLs. In fact, we still do not know how long the response to antibiotics lasts and, more importantly, we do not have reliable markers to distinguish between Cp persistence and reinfection in unresponsive cases. In the case of Cp persistence, a different schedule of antibiotic therapy should be considered or, alternatively, a new antibiotic could be implemented. At a variance with this, reinfection from an external source may occur more frequently than expected, as suggested by the demonstration of two metachronous Cp-associated lymphomas in the same patient [8]. These two lymphomas displayed a different histotype and site of origin (OAMZL and diffuse large B cell of the bronchus, respectively) and, most importantly, were clonally unrelated [8]. When a possible reinfection is suspected, the presence of an external reservoir for Cp should be carefully investigated in the domestic or professional environment. The pathogenic and clinical relevance of external sources of infection/reinfection is also corroborated by the finding that OAMZL patients reported prolonged contact with pets, as opposed to controls [4]. Assessment of the relationship with pets is, therefore, an urgent issue in Cp-related OAMZL, and well-designed epidemiological studies are required to clarify these preliminary observations.

Most studies carried out to assess the prevalence of bacterial infection in these lymphomas were mainly focused on the evaluation of the presence of Cp. To fulfill this aim, molecular detection of the 16S–23S spacer region and chlamydial Hsp60 genes were the main targets [3]. Although the PCR assays targeting these regions of the Cp genome proved to be useful for diagnostic purposes, they do not provide sufficient data concerning the molecular features of the Cp strain(s) infecting OAMZL patients. This is a crucial issue from a pathogenic standpoint, and relevant insights in this respect could derive from the characterization of the genotype of the Cp strains isolated from OAMZL. Particularly informative will be the detailed analysis of the outer membrane protein-A (OMP-A) sequences [9], which could allow the identification of immunogenic epitopes possibly triggering abnormal T-cell responses, and which may be useful for the development of immunomonitoring assays for clinical purposes.

Current studies have reported a wide range of prevalence rates for the association between Cp infection and OAMZLs. Although these studies were not carried out using homogeneous analytical methods, there is a generally accepted viewpoint toward the existence of geographic variability of these prevalence rates worldwide [1], a feature shared by most infection-driven tumors. A large study involving an adequate number of cases from several countries needs to be conducted with the same methodological diagnostic approach.

Another interesting point, although preliminary, is the clinical response to doxycycline observed in a proportion of patients with Cp-negative OAMZL [10]. This aspect is highly relevant, since it may suggest that Cp infection is below the levels of detection in these patients or, more appealingly, that additional bacteria responsive to these antibiotics other than Cp may be present in OAMZLs. Our group is presently investigating this hypothesis.

Chlamydophila psittaci infection has been historically associated with OAMZL. However, more recent data have provided evidence that this infection could also occur in other anatomical sites and in additional histological categories of lymphomas, although with a much lower prevalence rate. In particular, a near-significant association of intralesional Cp DNA with cutaneous and Waldeyer’s ring B-cell lymphomas has been reported in a small series of these tumors in a screening study [3]. Ad hoc studies on larger series are required to draw definitive conclusions in this respect. Along this line, the presence of Cp infection has been documented within the tumor tissue of clinically aggressive lymphomas, in particular diffuse large B-cell lymphomas (DLBCLs). One of these Cp-related DLBCLs has been treated with doxycycline as a first-line therapy, with complete remission being obtained [8]. This experience is presently anecdotal, but it should be seriously considered, since other situations, such as the exclusive antibiotic treatment of Helicobacter pylori-related gastric DLBCL, have attained an interesting rate of clinical response [11].

Finally, a promising unconventional therapeutic approach could be represented by a 6-month oral administration of clarithromycin, which has been tested in a Phase II trial including 12 patients with OAMZL; notably, the majority of these patients were previously treated with doxycycline. Clarithromycin has been successful in obtaining a 38% clinical response rate in these patients, and in particular in those with conjunctival disease [10].

In conclusion, the currently available treatment modalities for OALs are probably insufficient to cover the heterogeneous spectrum of the clinicopathological entities described. In addition, since the vast majority of these lymphomas are represented by OAMZLs, an upfront approach with systemic therapy and/or radiotherapy appears excessive. Antibiotic treatment appears to be a promising alternative source with a significant rate of success. Prospective, forthcoming studies will better contribute to dissect the ‘pros and cons’ of this novel strategy in OAMZL.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.