Original Article | Published:

Graft-versus-host Disease

Long-term safety and efficacy of autologous platelet lysate drops for treatment of ocular GvHD

Bone Marrow Transplantation volume 52, pages 101106 (2017) | Download Citation


Current ocular GvHD (oGvHD) treatments are suboptimal. We investigated the safety and efficacy of long-term continuous treatment with autologous platelet lysate (PL) drops in patients with oGvHD Dry Eye Syndrome (DES) score 2–3 refractory to topical conventional therapy. Ophthalmic evaluation was performed at 6 month intervals. Symptoms were assessed using the Glaucoma Symptom Scale (GSS). Patients were defined 'responders' when showing a reduction at least one grade on National Institutes of Health Eye Score from baseline at the 6 month visit. Thirty-one patients were included, and 16 (51%) completed 36 months of follow-up (range 6.5–72.7). At 6 months all patients were classified as responders: median GSS symptom score decreased from 70 to 41 (33 at 36 months), median GSS function score reduced from 68 to 46 (33 at 36 months) (all P<0.001). Median Tear Break Up Time improved from 3 to 6 s after 6 months and was maintained over time. All signs improved at 6 and 36 months (clinical and statistical significance). No severe adverse events occurred. Long-term treatment with PL drops is secure and effective for oGvHD and can be an efficient therapy option from initial stages of oGvHD to prevent permanent ocular impairment and improving quality of life.


Ocular (oGvHD) occurs in 60% of HSCT recipients1 and it is strongly associated with extensive chronic GvHD (cGvHD) (60–90%).2 Clinical manifestations of oGvHD are diverse, and can affect conjunctiva, lacrimal gland, cornea, lid and vitreous. The most frequent ocular manifestation of GvHD is Dry Eye Syndrome (DES), observed in 69–77% of patients with chronic systemic GvHD and characterized by keratoconjunctivitis sicca, conjunctival inflammation and chronic blepharitis.3

Though not life-threatening, oGvHD DES can greatly hinder patients’ quality of life and may be visually debilitating.1 According to National Institutes of Health (NIH) guidelines revised in 2014, DES is classified with a range from 0 (‘no symptoms‘) to 3 (‘severe dry eye symptoms significantly affecting daily living (special eyeware to relieve pain) or unable to work because of ocular symptoms or loss of vision due to keratoconjunctivitis sicca’).4

DES can rapidly progress to severe irreversible complications with an increased risk of corneal ulceration and occasional perforation leading to a permanent visual loss.5, 6, 7

Autologous serum (AS) eye drops were the first peripheral blood-derived products used in the treatment of corneal diseases (for example, Sjogren syndrome);8 in AS, growth factors such as EGF (epidermal growth factor), PDGF and FGF (fibroblast growth factor) mediate the corneal tissue repair process. However, AS may contain proinflammatory agents derived by leukocyte degranulation, which may hamper ocular tissue regeneration.9

Despite many advances in the therapy of oGvHD, safety and efficacy data in long-term treatment are lacking: to our knowledge, the longest follow-up reported in oGvHD patients is 1 year in a sample of 12 patients treated with topical tacrolimus ointment.10 Recently, our group demonstrated the clinical efficacy and safety of autologous platelet lysate (PL) eye drops in patients with oGvHD refractory to treatment with conventional therapy with 80% of responders after 1 month of treatment and maintained for 6 months.11

Based on our initial data, we designed the present study with the aim of evaluating the long-term continuous treatment with PL drops in patients affected by oGvHD refractory to conventional therapy. The main goal of our study was to evaluate the long-term safety and efficacy of a continuous treatment with PL eye drops in a sample of patients with moderate–severe oGvHD.

Materials and methods

This prospective cohort study was approved by our Ethics Committee and conducted at the teaching Hospital of Pavia from January 2011 to January 2015. All patients gave their written consent prior to inclusion, were enrolled between 2011 and 2013 and followed-up for up to 3 years.


Eligibility criteria were:

  • Age 18 years.

  • cGvHD with oGvHD showing score 2 or 3 of NIH Organ Scoring of Chronic GvHD for eyes. 4

  • Refractory to treatment with preservative-free artificial tears for at least 3 months.

  • Platelets in peripheral blood>120000/μL.

  • No systemic infection in process (including HIV, hepatitis B virus, hepatitis C virus) at time of enrollment and PL eye drops preparation.

  • No ophthalmic infections in process, retinal disease, ocular hypertension disease or other than DES (except cataract) at the time of recruitment.

  • Stable systemic and ocular topical therapy lasting at least 3 months.

  • No systemic or ocular surgery (including cataract surgery) in the 3 months preceding enrollment.

  • Able to read and complete dry eye symptoms questionnaire in the Italian version.

  • At least 6 months of continuous PL eye drops therapy.

  • Written consent.


The procedures for PL drops preparation was obtained by the same method as described in our previous report.11 In brief, 40 ml of anticoagulated peripheral blood were drawn from the patient and centrifuged to obtain platelet rich plasma. Afterwards, it was diluted with a saline solution and aliquoted in 1.5 ml sterile vials. Vials were frozen and thawed to induce PDGF release and stored in patients’ own freezers for 45 days. As described elsewhere,11 each vial was thawed and used for a maximum of 24 h, and PL eye drops were instilled 4 times/day. PL therapy was suspended in the case of conjunctivitis onset and re-started after appropriate local therapy. To exclude bacterial or fungal contamination of the stored PL eye drops, microbiological cultures of vials were obtained after their use. All patients were under PL treatment during follow-up.

Ophthalmological assessments and follow-up

The same Ophthalmologist evaluated all patients (to reduce clinical assessment variability) at the time of enrollment and every 180 days of treatment. Additional ophthalmological evaluations were also performed 30 and 90 days after enrollment and every 90 days after the first 6 months (data are not reported in this study).

Dry eye symptoms over time were assessed using the Glaucoma Symptom Scale (GSS).12 The GSS, the only questionnaire validated in Italian for symptoms of ocular surface disease, quantifies ocular symptoms including 10 ocular complaints, some of non-visual nature (burning/stinging, tearing, dryness, itching, soreness/tiredness, sensation of foreign body in the eye: ‘GSS Symptom score’) and some of visual nature (blurry vision, hard to see in daylight, hard to see in dark places, halos around lights: ‘GSS Function score’). For each eye, the questionnaire provides a score ranging from 0 to 100, with 100 representing worse symptoms/function and 0 absence of the issue. The final GSS score is an un-weighted average of the responses to all 10 items; the GSS subscale Symptom score (Symp-6) is an un-weighted average of all items that comprise the particular subscale (see Figure 1). GSS is well correlated to ocular signs (as lissamine staining, fluorescein staining and Median Tear Break Up Time (TBUT)13). At each visit, the Ophthalmologist recorded information on return to daily activities for each patient. Objective evaluations included best corrected visual acuity for distance, determination of tear film using the TBUT test (expressed as mean value after three measurements), anterior and posterior segment biomicroscopy, corneal fluorescein staining and conjunctival lissamine staining, both evaluated following the Oxford scheme.14 Corneal biomicroscopy was evaluated using a customized scale considering the percentage of extension of lesions on corneal surface (absent: no corneal lesions, <10%, 10–50%, >50%, ulcer). Schirmer test (without local anesthesia) was performed only at time of enrollment.4

Figure 1
Figure 1

GSS-Glaucoma symptom scale.

Patients were defined as ‘responders’ when they presented a reduction at least one grade on NIH eye score from baseline at the 6-month visit; ‘worsening’ was defined as an increase of NIH eye score from previous visit, after reaching response; ‘stability’ was defined as anything else (for example, same or inferior grade of NIH eye score at subsequent visit).

Statistical issues

Descriptive statistics were produced for all variables. Mean and s.d. are presented for normally distributed variables, and median and interquartile range (IQR) for non-normally distributed variables, number and percentages for categorical variables. For descriptive statistics, only the right eye is presented. Generalized linear models for repeated measures (patients and eyes) were applied to assess changes over time in symptoms and signs. In all cases, two-tailed tests were used. P-value significance cutoff was 0.05. The primary outcome measure of the study was GSS general score over time. Secondary outcome measures were GSS subscales (symptoms and function, for each eye separately and overall), and ophthalmological examination parameters (TBUT, fluorescein staining, lissamine staining and amount of corneal involvement). Stata computer software version 14.0 (Stata Corporation, 4905 Lakeway Drive, College Station, TX 77845, USA) was used for statistical analysis. Statistical code for analysis is available upon request.


Of the 31 patients enrolled (Table 1) 21 (68%) were males, with a mean age 47 years (range 33–60). One patient was only treated in the right eye owing to herpetic keratitis; therefore, the number of eyes included in the analysis is 61. At the time of enrollment (T0), all patients used artificial eye drops several times per day (median 10 (IQR 8–20)), median GSS symptom score was 70 (IQR 58–83), median of GSS function score was 68 (IQR 50–100) and median TBUT was 3 s (IQR 1–5).

Table 1: Characteristics of patients with chronic ocular GvHD enrolled

Median follow-up was 36 months (range 6.5–72.7). Five patients (16%) died for hematological relapse, one patient (3%) died for GvHD complication, one patient dropped out for other than eye GvHD complications and two for logistic reasons. Sixteen patients (51%) completed 36 months of follow-up. Throughout follow-up, systemic immunosuppression was unchanged in 26 (84%) patients, and reduced in five (16%) patients: one patient withdrew cyclosporine and steroids, two patients withdrew steroids, one patient withdrew cyclosporine but continued steroids, one patient withdrew cyclosporine (that is, his whole therapy). Overall five patients died within the first 36 months (three owing to disease relapse, one patient of poliserositis with sudden death, one of sepsis). One additional patient died at 40 months (of disease relapse).

Description of treatment and ocular signs/symptoms over time are shown in Table 2. In brief, at 6 months median GSS symptom score fell to 41 (IQR 25–50), median GSS function score was 46 (IQR 31–56), the median number of ocular instillations for day was reduced to 4 (IQR 3–6); and median TBUT improved to 6 s (IQR 5–9) (all P<0001) (Table 2). At the 6-month visit, all patients were defined ‘responders’: all improved in their GSS symptoms score and at least one sign (all improved at least 1 s longer TBUT, and showed a reduction of at least 1 point on the Oxford lissamine scale classification). Twenty patients (66%) had a reduction of Oxford fluorescein corneal staining 2, and all patients had a reduction at least one point on corneal extension grade scale. Only two patients showed temporary relapse in symptoms and signs after improvement; in both cases ophthalmologic relapse was associated to worsening systemic cGvHD (one with lung involvement and one with skin involvement). Only 13 episodes in 10 patients of temporary worsening of ocular symptoms were observed: six patients had symptom worsening owing to cataract onset (difficulty in night vision n=4 and photophobia n=2), four patients had foreign body feeling in the eye (owing to recent cataract surgery), three patients had burning sensation owing to a viral conjunctivitis’. After the 3-year follow-up, GSS symptom and function score further reduced to a median 33 (IQR 0–41) and 33 (IQR 0–56), respectively, median of ocular instillations for day was unchanged from month 6, and median TBUT was stable at 6 s (IQR 7–9) (all P<0.001). Oxford lissamine grade was unchanged, 80% of patients presented a fluorescein staining lower than two and corneal extension grade improved in all but one patients at least 1 point from month 6. In Figure 2 we report the average General GSS over time for the whole group.

Table 2: Description of treatment and ocular signs/symptoms over time
Figure 2
Figure 2

Reduction of symptoms over time, in terms of GSS general score.

Systemic immunosuppression was unchanged (87%) or decreased (13%) and systemic treatment had not been intensified in any patient in the follow-up period.

Optimal compliance to PL eye drops was observed in all patients. No patient stopped the treatment for complications or adverse events, but two patients, after reaching response, dropped out for logistical reasons (they lived 760 and 1500 km from our hospital). No fungal or bacterial contamination was noticed in plasma rich in PDGFs eye drop samples, and no contamination developed during storage.


Despite a still high all-cause mortality rate, advances in hematological therapies allow prolonged survival in patients who undergo HSCT.15 Therefore, long-term strategies in managing GvHD manifestations and improving quality of life are pivotal. The aim of oGvHD treatment is to avoid irreversible ocular complications (in particular DES) that, in turn, affect daily quality of life.3 Our current results, based on our initial experience in short-term (6 months) treatment of patients with oGvHD with autologous PL eye drops11 indicate the safety and efficacy of autologous PL eye drops in a longer follow-up. In fact, despite moderate–severe DES (grade 2–3) at enrollment, all patients responded, showing improvement in signs and in symptoms at 6 months; importantly, this improvement was still evident after 3 years of PL therapy. Our results confirm that PDGFs have a crucial role in preserving corneal surface, being involved in the renewal of corneal epithelium and wound-healing processes by interaction with the different corneal cellular components.16 Platelets store a range of potent bioactive mediators, including chemokines and PDGFs in different types of isoforms, TGF β (transforming growth factor β), insulin like growth factors-1, VEGF (vascular endothelial growth factors), EGF, bFGF (basic fibroblast growth factor) and others. When platelets are activated they release their granule contents: proinflammatory cytokines and chemokines, PDGFs, TGF β that contribute to wound healing by attracting macrophages and granulocytes to remove damaged cells and tissue, activating proliferation of fibroblasts and endothelial cells. EGF, bFGF, VEGF promote granulation tissue, collagen synthesis and angiogenesis with a wound contraction. In particular, the release of these pools of mediators by platelets promote re-epithelization.17

Compared with AS, PL contains a significantly higher level of growth factors because it is obtained from platelets, which are richer in growth factors, whose concentration is further increased by the freeze–thawing protocol inducing PL. In particular, the higher level of EGF, enhancing proliferation and migration on corneal epithelial cells, accelerates the wound-healing. Furthermore, the significant higher level of FGF, TGF β and PDGF detected in PL eye drops may explain the enhanced proliferation and migration activities of fibroblasts when compared with AS. In addition, the higher efficacy of PL than AS may be due to the higher concentration of PDGFs, and to the absence of contaminating leukocytes.9

All patients improved during PL treatment. In fact, even those two patients with worsening GvHD in other organs experienced an improvement in ocular GvHD. This is not surprising, considering that the eye is an immunological sanctuary, not necessarily associated to the global GvHD tendency (for instance, ocular GvHD does not always respond to the systemic immunosuppressive therapy prescribed for extra-ocular cGvHD and poorly responds to the immunomodulation of extracorporeal photopheresis treatment).

Despite the fact that responders did not obtain complete normalization of symptoms and signs, improvement was sufficiently marked to provide significant improvement of patients’ quality of life and return to daily life activities. Of note, all patients returned to their pre-transplant daily activities (within the limits imposed by the systemic chronic condition of GvHD). Interestingly, patients’ compliance was very high, and even those residing up to 1500 km requested to continue PL eye drops therapy at our center where this option is available.

The only episodes of temporary ocular worsening were due to cataract onset or viral conjunctivitis. In the former case, patients underwent cataract extraction and after surgery the corneal surface worsened for a few weeks. In the latter, viral conjunctivitis was always associated with systemic viral infection, and the viral origin was confirmed by conjunctival swabs, which also excluded bacterial or fungal contamination of PL eye drops.

Also, when prepared in-house at a public Institution, autologous PL treatment is feasible at a relatively low cost. In our region (Lombardy-Italy) in 2016, the reimbursed costs are 202 Euros for each lot (that is, 30 samples) of PL eye drops.

Good short-term clinical results have been obtained with other topical therapies. Lubricants have shown a global amelioration no greater than 25% from baseline;18 steroids are frequently used, but long-term use can cause undesirable side effects, such as steroid-induced ocular pressure elevation, cataract onset and superinfection;19 ocular topical cyclosporine induced an improvement in 69% of 13 patients after 2 months of follow-up;20 bandage and scleral lenses obtained prompt improvement in 50% of 20 patients,21, 22, 23 tacrolimus demonstrated good safety and efficacy in 13 patients in a mean of 12.2 months;10 finally, AS eye drops are secure and efficacious after 4 weeks of treatment in the majority of patients.24

In a longer-term follow-up, only ocular topical tacrolimus ointment used for 12 months and was shown to be effective;10 however, episodes of blurred vision or a burning sensation were described during treatment. In fact, other studies suggest that ocular topical tacrolimus has an increased risk of human herpes simplex keratitis.25, 26 Food and Drug Administration guidelines underline that ocular topical tacrolimus should be utilized only for short periods and it should be avoided in patients with compromised immune systems.

We must acknowledge some limitations of our study. First, the restricted sample does not allow more in-depth analyses on factors favouring response. Furthermore, neither patients nor the clinician were blind to the therapy. Also, there was no control group, because corneal condition can worsen rapidly in patients with cGvHD, and it may be difficult to select control subjects to be treated with conventional artificial eye drops. Finally, as per NIH guidelines, we assessed Schirmer’s test only at the time of enrollment, as it is not useful for follow-up of oGvHD, owing to poor correlation with symptoms change.4

To our knowledge, this is the first study using PL in the long-term in the field of oGvHD with successful results in terms of objective improvement (TBUT, fluorescein and lissamine staining, corneal damage) and vision-related quality of life as defined by symptom reduction without relevant side effects. Our data, therefore, suggest that PL eye drops treatment can be an effective and safe therapeutic option for oGvHD. Prompt diagnosis and adequate management of ocular manifestations of GvHD can avoid irreversible complications, improve quality of life, and allow patients to return to pre-transplant daily activities.


  1. 1.

    , , , , , . Risk factors for the development of ocular graft-versus-host disease (GVHD) dry eye syndrome in patients with chronic GVHD. Br J Ophthalmol 2015; 99: 1514–1518.

  2. 2.

    , , , , , . Comparison of stem cell sources in the severity of dry eye after allogeneic haematopoietic stem cell. Br J Ophthalmol 2012; 96: 34–37.

  3. 3.

    , , , , , et al. A cross-sectional study on vision-related quality of life in patients with ocular GVHD. Bone Marrow Transplant 2015; 50: 1224–1226.

  4. 4.

    , , , , , et al. National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: I. The 2014 Diagnosis and Staging Working Group report. Biol Blood Marrow Transplant 2015; 21: 389–401.

  5. 5.

    , , , , , et al. Grading criteria for chronic ocular graft-versus-host disease: comparing the NIH eye score, Japanese dry eye score, and DEWS 2007 score. Sci Rep 2014; 4: 6680.

  6. 6.

    , , , , , et al. Predictive models for ocular chronic graft-versus-host disease diagnosis and disease activity in transplant clinical practice. Haematologica 2015; 100: 1228–1236.

  7. 7.

    , , , , , et al Ocular complications in hematopoietic stem cell transplant patients-a 10 year experience at Mayo Clinic. 2016. EBMT 42nd Annual Meeting.

  8. 8.

    , , , , , et al. Autologous serum application in the treatment of neurotrophic keratopathy. Ophthalmology 2004; 111: 1115–1120.

  9. 9.

    , , , , , . Plasma rich in growth factors (PRGF) eye drops stimulates scarless regeneration compared to autologous serum in the ocular surface stromal fibroblasts. Curr Pharm Biotechnol 2012; 13: 1257–1265.

  10. 10.

    , , , , , . Long-term result of maintenance treatment with tacrolimus ointment in chronic ocular graft-versus-host disease. Am J Ophthalmol 2015; 159: 519–527.

  11. 11.

    , , , , , . Autologous platelet lysate for treatment of refractory ocular GVHD. Bone Marrow Transplant 2012; 47: 1558–1563.

  12. 12.

    , , , , , et al. The Italian version of the Glaucoma Symptom Scale Questionnaire: translation, validation, and reliability. J Glaucoma 2013; 22: 44–51.

  13. 13.

    , , , . Symptoms Ocular surface disease and glaucoma: how to evaluate impact on quality of life. J Ocul Pharmacol Ther 2013; 29: 390–394.

  14. 14.

    , , . Grading of corneal and conjunctival staining in the context of other dry eye tests. Cornea 2003; 22: 640–650.

  15. 15.

    , . Ocular manifestations of graft-versus-host disease: 10 years' experience. Clin Ophthalmol 2015; 9: 1209–1213.

  16. 16.

    , , , , , et al. Platelet lysate formulations based on mucoadhesive polymers for the treatment of corneal lesions. J Pharm Pharmacol 2010; 63: 189–198.

  17. 17.

    , , , . Human platelet lysate: Replacing fetal bovine serum as a gold standard for human cell propagation? Biomaterials 2016; 76: 371–387.

  18. 18.

    , . Efficacy of different dry eye treatments with artificial tears or ocular lubricants: a systematic review. Ophthalmic Physiol Opt 2009; 29: 573–583.

  19. 19.

    , , , , , et al. Topical corticosteroid therapy for cicatricial conjunctivitis associated with chronic graft-versus-host disease. Bone Marrow Transplant 2004; 33: 1031–1035.

  20. 20.

    , , . High-frequency topical cyclosporine 0.05% in the treatment of severe dry eye refractory to twice-daily regimen. Cornea 2009; 28: 1091–1096.

  21. 21.

    , , , , , et al. Bandage soft contact lenses for ocular graft-versus-host disease. Biol Blood Marrow Transplant 2015; 21: 2002–2007.

  22. 22.

    , , , , . Bandage and scleral contact lenses for ocular graft-versus-host disease after allogeneic haematopoietic stem cell transplantation. Acta Ophthalmol 2015; 93: e604.

  23. 23.

    , , , , , et al Sclero-corneal lenses safe and efficient for the treatment of keratoconjunctivitis sicca in patients with refractory ocular GVHD: a study on be half of the SFGM-TC. 2016 EBMT 42nd Annual Meeting.

  24. 24.

    , , , , , et al. Autologous serum eye drops for the treatment of severe dry eye in patients with chronic graft-versus-host disease. Bone Marrow Transplant 2003; 31: 579–583.

  25. 25.

    , , , , . Therapeutic effect of 0.03% tacrolimus ointment for ocular graft versus host disease and vernal keratoconjunctivitis. Korean J Ophthalmol 2012; 26: 241–247.

  26. 26.

    , , . Topical tacrolimus ointment for treatment of refractory anterior segment inflammatory disorders. Cornea 2005; 24: 417–420.

Download references

Author information


  1. Department of Ophthalmology, University Eye Clinic of Pavia, IRCCS Policlinico S Matteo Foundation, Pavia, Italy

    • S Pezzotta
    • , G C Rossi
    • , P E Bianchi
    •  & E Antoniazzi
  2. Immunohaematology and Transfusion Service, Apheresis Unit, IRCCS Policlinico S Matteo Foundation, Pavia, Italy

    • C Del Fante
    •  & C Perotti
  3. Biostatistic and clinical epidemiology unit, Scientific Direction, IRCCS Policlinic San Matteo Foundation, Pavia, Italy

    • L Scudeller


  1. Search for S Pezzotta in:

  2. Search for C Del Fante in:

  3. Search for L Scudeller in:

  4. Search for G C Rossi in:

  5. Search for C Perotti in:

  6. Search for P E Bianchi in:

  7. Search for E Antoniazzi in:

Competing interests

The authors declare no conflict of interest.

Corresponding author

Correspondence to S Pezzotta.

About this article

Publication history







Further reading