Article Text

Download PDFPDF
Three-year OCT predictive factors of disease recurrence in eyes with successfully treated myopic choroidal neovascularisation
  1. Enrico Borrelli1,2,
  2. Marco Battista1,2,
  3. Giovanna Vella1,
  4. Riccardo Sacconi1,2,
  5. Lea Querques1,
  6. Domenico Grosso1,2,
  7. Francesco Bandello1,2,
  8. Giuseppe Querques1,2
  1. 1 Department of Ophthalmology, IRCCS Ospedale San Raffaele, Milan, Italy
  2. 2 Department of Ophthalmology, University Vita-Salute San Raffaele, Milan, Italy
  1. Correspondence to Professor Giuseppe Querques, Ophthalmology, Ospedale San Raffaele, Milano, Italy; giuseppe.querques{at}


Purpose To assess the relationship of demographics, clinical characteristics and structural optical coherence tomography (OCT) findings to disease recurrence in a cohort of patients with newly diagnosed myopic choroidal neovascularisation (CNV)

Methods In this retrospective, longitudinal study, a total of 64 participants (64 eyes) with successfully treated myopic CNV had obtained resolution of exudation after treatment (study baseline) and with 3 years of regular follow-ups. Several baseline OCT qualitative features and quantitative measurements were assessed at baseline and included in the analysis. Main outcome measures included incidence of disease recurrence and HR for demographics, clinical characteristics and OCT risk factors.

Results At month 36, 40 eyes (62.5%) developed disease recurrence (active CNV). Multivariate linear regression analysis revealed that final visual acuity (dependent variable) was associated with visual acuity at the first visit after complete resolution of exudation (p<0.0001), baseline size of patchy atrophy (p=0.010), baseline subfoveal choroidal thickness (p=0.008), baseline maximum CNV height and width (p=0.011 and p=0.003) and recurrence of CNV exudation (p=0.007). The following factors were associated with an increased risk of disease recurrence: size of patchy atrophy had an HR of 1.14 (95% CI 1.01 to 1.29; p=0.036); maximum CNV width had an HR of 1.02 (95% CI 1.01 to 1.04; p<0.0001).

Conclusion We identified OCT risk factors for the disease recurrence in eyes with successfully treated myopic CNV. Assuming that disease recurrence is a sight-threatening event, our findings may help in the identification of high-risk patients and eventually ameliorate their outcome.

  • imaging
  • macula
  • neovascularisation

Data availability statement

Data are available upon reasonable request. Data are available upon request to the corresponding author.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Myopia is a common condition worldwide with an estimated prevalence ranging from 25% in the 1USA and Europe2 to 40% in East Asia.3–5 The term myopic macular degeneration (also known as pathologic myopia, myopic maculopathy or degenerative myopia) refers to the presence of distinctive features including myopic choroidal neovascularisation (CNV), chorioretinal atrophy and lacquer cracks.6 7 Myopic macular degeneration has an estimated prevalence of 3% in the global population8 and this entity was demonstrated to be more common in eyes with spherical equivalent refractive errors equal to or superior to −6 diopters (D) and axial lengths greater than 26.5 mm.9

CNV is a frequent complication in myopia as it was estimated to affect up to 5%–11% of individuals with myopic macular degeneration.8 More importantly, myopic CNV represents a sight-threatening complication that may be associated with significant visual impairment.10 Furthermore, myopic CNV may affect the working-age population and, therefore, it may be associated with a considerable social and economic burden.10

Myopic CNV is a complex disease with multifactorial aetiologies. However, strong evidences suggest that this complication may be mainly secondary to a progressive and excessive ocular elongation with reduced choroidal flow,10 11 leading to a disproportion between proangiogenic and antiangiogenic factors.12 13

Previous studies have proved that the anti-vascular endothelial growth factor (VEGF) therapy is the current standard of treatment in newly diagnosed myopic CNV.14 These lesions are usually characterised by a rapid response to treatment and a fewer injections of anti-VEGF are usually required to resolve exudation, in comparison with CNV associated with other conditions (eg, age-related macular degeneration).15 However, after exudation is resolved, patients may experience recurrences that were demonstrated to significantly impact the long-term visual outcomes.16 17 Conversely, early diagnosis and prompt intervention in eyes with myopia and active CNV have been proved to ameliorate visual outcomes.18 Therefore, it would appear to be of great relevance to detect recurrences at the earliest stage possible. More importantly, the identification of risk factors for the development of recurrences may expand our knowledge regarding their pathogenesis.

Structural optical coherence tomography (OCT) is an essential diagnostic tool for the evaluation of subjects with myopia as it provides morphological information about the retinal and choroidal layers. Several previous reports have distinguished and described OCT findings associated with myopic CNV.19–22 On structural OCT, myopic CNV has the typical characteristics of a ‘type 2’ pattern of neovascularisation as it is visualised as a hyperreflective material under the neuroretina and above the retinal pigment epithelium (RPE). In general, minimal signs of exudation are associated, consisting of presence of intra/subretinal fluid and/or fuzziness of the CNV border.19 20 After resolution of exudation, structural OCT usually demonstrates a reduction in size of the subretinal hyperreflective lesion with disappearance of the fuzzy margin and sub/intraretinal fluid.19 20 Importantly, structural OCT was revealed to be a sensitive imaging modality to distinguish active versus nonactive CNV lesions.21 22

In this longitudinal study over 3 years of follow-up, we examined the relationship of demographics, clinical characteristics and structural OCT findings to the development of exudation recurrences in a cohort of eyes with newly diagnosed myopic CNV successfully treated with anti-VEGF therapy.


The San Raffaele Ethics Committee was notified about this retrospective cohort study. Based on the Italian legislation, this kind of study does not require the Ethics Committee approval, as it is required only a notification. The study adhered to the 1964 Helsinki declaration and its later amendments. Informed consent was obtained from all individual participants included in the study.


In this retrospective longitudinal cohort study, subjects 18 years of age and older with newly diagnosed treatment-naïve exudative myopic CNV in at least one eye were identified from the medical records of a medical retinal practice (Medical Retina and Imaging Unit) at the San Raffaele Scientific Institute. Diagnosis of exudative myopic CNV was determined by clinical examination, structural OCT and fluorescein angiography (FA) evaluation, as previously described.10 After receiving a diagnosis of treatment-naïve active myopic CNV, patients received a single injection of anti-VEGF (aflibercept or ranibizumab) followed by an as-needed reinjection retreatment strategy based on the OCT and FA findings. Therefore, after diagnosis, patients underwent a variable number of injections (at least one) until CNV activity was resolved. Subjects were included in the initial study cohort if they had resolution of exudation after anti-VEGF treatment in at least one follow-up visit, which was considered as the inclusion baseline visit in our study.

Exclusion criteria at baseline visit included: (1) evidence of myopic traction maculopathy23, (2) evidence of inflammatory disorders, including multifocal choroiditis and punctate inner choroidopathy, that are known to be potential causes of CNV in myopic eyes24, (3) history of vitreoretinal ocular surgeries, (4) history of any ocular surgeries during the follow-up period (36 months after the baseline visit) and (5) history or evidence of other retinal and optic nerve disorders. The population fulfilling these characteristics was the starting cohort for this analysis (n=512) (online supplemental figure 1).

To be included, subjects were also needed to have a minimum of 4 yearly visits with OCT imaging obtained and covering a study period of 3 years (36 months) after the baseline visit. Of note, all subjects at our clinic are instructed to attend visit urgently in the event of changes in visual symptoms and/or visual decline. Subjects were also educated to perform Amsler grid test with regularity. After this further evaluation, 88 subjects were potentially includable in our analysis (online supplemental figure 1).

Structural spectral domain OCT and combined near-infrared reflectance (NIR) imaging were performed with the Heidelberg Spectralis Heidelberg Retina Tomograph (HRT)+OCT device (Heidelberg Engineering, Heidelberg, Germany). The OCT images were acquired with two standardised acquisition protocols, as follows: (1) six radial linear OCT B-scans with the enhanced depth imaging (EDI) technique centred on the fovea, each composed by 25 averaged OCT B-scans (768 A-scans per line) at 30 degrees and (2) 19 horizontal OCT B-scans, each of which comprised 25 averaged scans, covering approximately a 5.5×4.5 mm area centred on the fovea. Additional high-resolution single scans were performed in specific regions if necessary (eg, in those cases in which conventional scans were not able to capture the entire CNV lesion, additional scans were performed as guided by the CNV location/extension). A minimum signal strength of 25 was required to the OCT images to be included, as recommended by the manufacturer.25 At each follow-up visit, structural OCT scans were performed at the same location using the follow-up function. In addition, at each visit, enrolled patients had a complete ophthalmologic assessment, including the measurement of best-corrected visual acuity (BCVA), dilated fundus examination, blue fundus autofluorescence and FA in those cases with suspected active CNV in which structural OCT was not diriment (this decision was made by the attending physician). The Heidelberg Spectralis HRA+OCT device was used to perform dye angiography. The refractive error was expressed in spherical equivalent.26 In subjects with a history of cataract surgery, preoperative refractive errors were considered in the analysis.

OCT grading

Two independent and experienced readers (EB and MB) first reviewed structural OCT images for eligibility (88 subjects). In detail, OCT images at baseline were graded and eyes not fulfilling the study inclusion and exclusion criteria were excluded. Furthermore, eyes with signs of CNV activity at the inclusion visit were excluded from the analysis (or a different inclusion visit was selected). After applying these criteria, 64 subjects were eligible for further analysis (online supplemental figure 1).

Therefore, eligible eyes (n=64) were independently graded for qualitative features and quantitative measurements by the same graders who were masked as to the final eyes’ outcomes.

In detail, OCT images at baseline were graded for qualitative features previously proposed as suggestive of RPE or choroidal distress in myopic eyes, as follows (online supplemental figure 2):

  • Patchy atrophy: this was graded as being present on the basis of previous reports.27 28 In detail, OCT images were graded for the presence of regions of patchy atrophy, which are seen as sharply areas with a loss of most layers of choroid, of the RPE and of the outer retinal layers. The peripapillary atrophy area was excluded from the analysis.

  • CNV-related macular atrophy: OCT images were also reviewed for the presence of perilesional regions of RPE atrophy, as previously described.29

Furthermore, the two graders assessed the CNV location that was categorised as being subfoveal or nonsubfoveal on the basis of preceding studies.16 17

Based on the previous literature,30 31 a lesion was graded as present if the grader had a more than 90% confidence that it was recognisable in at least one B-scan. Discrepancies between graders were resolved by additional discussion and open adjudication to yield a single assessment for each case. The final decision was made by the senior author (GQ) in those cases in which the two graders did not agree on a single consensus result.

Images at baseline were also graded for the following quantitative measurements (online supplemental figure 2):

  • Size of patchy atrophy: the graders outlined the border of each region of patchy atrophy on NIR images by combining structural OCT information. Regions of RPE atrophy are usually visualised as hyperreflective patches on NIR images because of the absence of blockage from the RPE and increased reflection of light from the sclera.32 The built-in software was employed to perform this measurement. In detail, regions of patchy atrophy were delineated on en face NIR images only if regions were confirmed to be areas of atrophy on structural OCT.

  • Subfoveal choroidal thickness: this assessment was performed on the horizontal radial OCT B-scan obtained with EDI through the fovea, as previously described in myopic eyes.33

  • CNV size: the built-in software callipers were used to measure the maximum lesion height and width on OCT images.

For each measuring, the average value between the measurements performed by the two graders was finally included in the analysis.

Finally, the set of follow-up visits was graded for recurrence of CNV exudation by combining reports by attending physicians with multimodal imaging data.10 19–22

Study outcomes and statistical analysis

The primary outcome measure was the recurrence of CNV activity at any of the follow-up visits. A multivariate Cox regression model was fit to ascertain whether the baseline structural OCT characteristics in these eyes and other clinical/demographic characteristics (ie, age, gender, fellow eye status, refractive error, number of anti-VEGF injections to resolve exudation before the baseline visit) were risk factors for the development of macular complications. HRs and 95% CIs were computed. For continuous variables, HR estimates were calculated to furnish the increase (or decrease) in hazard corresponding to an X-unit rise in the quantification of the explanatory variable. The selection of X was set depending on the unit of measurement and the range of observed data values and had no influence on the significance of the explanatory variable. Patients with no recurrence detected throughout follow-up visits were censored at the last assessment (36-month visit).

Multivariate linear regression analysis was used to identify the factors that associate significantly with the final BCVA.

The unweighted k statistic test34 was performed to evaluate the agreement between graders in the assessment of OCT qualitative features at baseline and the study outcome of no evolution versus development of activity recurrence. The intraclass correlation coefficient (ICC) was calculated to examine the interreader agreement in terms of quantitative measurements.

Statistical calculations were performed using Statistical Package for Social Sciences (V.20.0, SPSS, Chicago, Illinois, USA).


Demographics and clinical characteristics of this study cohort are summarised in table 1. Mean±SD number of anti-VEGF injections before the inclusion visit was 2.1±1.3.

Table 1

Characteristics of patients included in the analysis

Among the cohort of 64 successfully treated myopic CNV at baseline, which were finally included in our analysis from the initial cohort of 512 participants (online supplemental figure 1), 40 eyes (62.5%) developed recurrence within 36 months. In details 33 eyes (51.6%) developed recurrence at the same location, whereas seven eyes (10.9%) had recurrence at a different location within 36 months. Recurrences occurred after 18.8±15.3 months.

Overall, the BCVA was 0.51±0.34 LogMAR (Snellen visual acuity (VA) of ~20/63) at the diagnosis of treatment-naïve exudative myopic CNV and 0.31±0.27 LogMAR (Snellen VA of ~20/40) at the first visit after complete resolution of exudation (baseline visit in our study; p<0.0001 with paired sample t-test). Mean±SD BCVA was 0.39±0.46 LogMAR (Snellen VA of ~20/50) at the last follow-up visit (p=0.105 with paired sample t-test in the comparison with the baseline visit’s values).

Regression analysis with final BCVA as the dependent variable

Results of multiple regression analysis are summarised in table 2. Final BCVA was associated with BCVA at the first visit after complete resolution of exudation (p<0.0001), baseline size of patchy atrophy (p=0.010), baseline subfoveal choroidal thickness (p=0.008), baseline maximum CNV height and width (p=0.011 and p=0.003) and recurrence of CNV exudation (p=0.007).

Table 2

Multiple regression analysis between final visual acuity (dependent variable) and other parameters

Baseline risk factors for the development of macular complications

At the baseline visit, a total of 32 eyes (50.0%) showed regions of patchy atrophy. Regions of CNV-related macular atrophy were found in 24 eyes (37.5%). Finally, the CNV was graded to be nonsubfoveal in 38 eyes (59.4%) (table 1).

In the quantitative grading, the average size of regions of patchy atrophy was 3.8±6.4 mm2. Subfoveal choroidal thickness was 65.2±45.5 µm. Mean±SD maximum CNV height and width were 126.5±65.1 µm and 726.9±442.8 µm, respectively (table 1).

Figure 1 summarises the results of the multivariate Cox regression analysis performed on the baseline OCT features as well as on the demographic and clinical characteristics at baseline, to determine the risk of exudation recurrence within 36 months (3 years). The following factors were associated with an increased risk: size of patchy atrophy had an HR of 1.14 (95% CI 1.01 to 1.29; p=0.036); maximum CNV width had an HR of 1.02 (95% CI 1.01 to 1.04; p<0.0001) (figures 2 and 3).

Figure 1

Graph showing the risk of disease recurrence. Graph shows the HR and 95% CI for the demographics, clinical characteristics and baseline OCT features. CNV, choroidal neovascularisation; OCT, optical coherence tomography; VEGF, vascular endothelial growth factor.

Figure 2

Multimodal imaging of a patient with successfully treated myopic choroidal neovascularisation (CNV) and developing disease recurrence. (A, B) Early and late-phase fluorescein angiography reveals an hyperfluorescent region (red arrow) with leakage corresponding to myopic CNV. (C) Structural optical coherence tomography (OCT) image confirms the presence of an active myopic CNV (white arrow) with fuzzy margin (visual acuity of 20/80 Snellen). (D) After two anti-vascular endothelial growth factor injections, structural OCT shows resolution of exudation (baseline visit in our study) (visual acuity of 20/32 Snellen). Furthermore, baseline risk factors for disease recurrence are visualised, including a large region of patchy atrophy and a wide inactive lesion. At the 14 month follow-up visit, (E) structural OCT confirms the presence of a disease recurrence with the CNV lesion that is characterised by an increased thickness and fuzzy margin (visual acuity of 20/63 Snellen).

Figure 3

Multimodal imaging of a patient with successfully treated myopic choroidal neovascularisation (CNV) and not developing disease recurrence. (A, B) Early and late-phase fluorescein angiography shows an area of hyperfluorescence (red arrow) with late leakage corresponding to a myopic CNV. (C) Structural optical coherence tomography (OCT) image demonstrates the presence of an active myopic CNV (white arrow) with fuzziness of margins (visual acuity of 20/25 Snellen). (D) After one anti-vascular endothelial growth factor injection, the CNV was small and without exudation (visual acuity of 20/20 Snellen). At the 36 month follow-up visit, (E) structural OCT is still characterised by the absence of signs of exudation (visual acuity of 20/20 Snellen).


The unweighted k values for intergrader repeatability were 0.86 (103/88) for eligibility criteria, 1.0 (64/64) for presence of patchy atrophy and 0.86 (60/64) for the presence of CNV-related macular atrophy. Agreement was reached for all discrepancies after adjudication between graders.

With regards to the manual measurements of subfoveal choroidal thickness, area of patchy atrophy and height and width of CNV dimensions, the interobserver agreements were good (ICCs were 0.910, 0.957, 0.945 and 0.932, respectively).


In this longitudinal investigation, we explored baseline predictors of exudative recurrence in patients with successfully treated myopic CNV. Overall, we reported OCT risk factors for developing exudative recurrences (ie, maximum CNV lesion width and size of patchy atrophy) within 36 months of follow-up. Among the demographic and clinical factors, age, gender and number of anti-VEGF injections to resolve exudation at the first occurrence did not significantly impact the 3-year expectation to have exudative recurrence.

Disease recurrence is a potential complication in eyes successfully treated for myopic CNV with an estimated prevalence ranging between 20.0% and 57.1%.10 16 17 Consistently, in our study cohort of eyes with resolved exudation at baseline after a mean number of 2.1±1.3 intravitreal anti-VEGF injections, the exudation recurrence was experienced by approximately 62% of patients at 3 years of follow-up (the average time before recurrence was 18.8±15.3 months). Several previous reports demonstrated that the occurrence of disease recurrence was significantly associated with worse visual acuity at the folast visit of follow-up.16 17 Our study confirms that these previous findings as we demonstrated that eyes developing recurrence were featured by a worse visual acuity at the final visit. Importantly, the present study highlights the distinctive relationship between visual acuity at 3 years and disease recurrence: considering final visual acuity as dependent variable in a multiple regression analysis, we observed a significant relationship between these two variables, even after accounting for other factors. The latter aspect further highlights the significance of distinguishing potential imaging risk factors for the development of disease recurrence.

In this study, the maximum CNV width measured on structural OCT images at baseline was associated with a greater risk for disease recurrence. Using structural OCT and fluorescing angiography, Yang and colleagues16 analysed 89 patients with a diagnosis of myopic CNV and a mean follow-up of 44.1 months. The authors of this study measured the baseline CNV size on FA images and correlated this value and other factors with disease recurrence within the follow-up period. Using a multivariate analysis, they demonstrated that baseline CNV size was the only significant risk factor for recurrence. Regardless, our analysis corroborated these findings and perhaps provides increased certainty in this presumed risk factor, given the high sensitivity of structural OCT for measuring CNV. In contrast to FA that allows a two-dimensional assessment, structural OCT grants a three-dimensional evaluation and, therefore, provides a more precise representation of the baseline characteristics that are risk factors for the development of disease recurrence. Finally, these results may support the theory that the CNV size may indirectly reflect the extension of choroidal flow abnormalities, the latter driving the pathogenesis of CNV, as previously suggested.16

In addition, this analysis identified a new baseline risk factor for disease recurrence. In detail, a larger region of patchy atrophy was identified on NIR and OCT imaging as a risk factor for disease recurrence. In our analysis, patchy atrophy was defined as loss of outer retinal layers and RPE and almost all of the choroidal layers on structural OCT images.27 28 Patchy atrophy was identified using structural OCT and its dimension was graded by integrating NIR imaging, which represents an en face modality that is often captured in combination with structural OCT. Patchy atrophy represents a recognised risk factor for myopic CNV. In a study of 218 consecutive myopic patients, Ohno-Matsui et al 35 investigated anatomical biomarkers that correlated with development of myopic CNV within the follow-up period. In the latter study, CNV developed in 20.0% of eyes with patchy atrophy at baseline. The authors concluded that patchy atrophy is a predisposing factor for the development of myopic CNV. Our analysis revealed that size of regions of patchy atrophy, rather than its presence, was a risk factor for disease recurrence within 3 years. As noted above, a myopia-related choroidal stretching may result in an overall ischaemia of the unit comprised of outer retina and RPE, this eventually culminating in raised VEGF levels in the sub-RPE compartment.10–13 Therefore, outer retinal ischaemia has been proposed as a mechanism driving the development of both patchy atrophy and CNV in myopic eyes. Thus, it is perhaps not surprising that a larger atrophy is associated with a higher recurrence rate as both these conditions may reflect a choroidal ischaemia in these eyes. Alternatively, it has been suggested that the RPE and Bruch’s membrane may be mechanically damaged at the edge of regions of patchy atrophy, and this may predispose to CNV development and recurrence.35 As stated above, the presence of RPE atrophy was not a significant risk factor for the development of recurrences. At least in part, this might be secondary to the relatively small sample size of our series. Future studies with larger sample sizes will clarify this aspect.

Importantly, we included one eye for each subject in order to examine the fellow eye status as a risk factor for disease recurrence. In our analysis employing multivariate model, baseline presence of myopic CNV in the fellow eye was not associated with a higher risk of disease recurrence within 3 years. Noteworthy, we also investigated other demographic and clinical factors that did not significantly impact disease recurrence.

Our study has limitations that must be considered in the evaluation of our findings. First, our study cohort was not part of a large multicenter trial and included subjects were not followed up at regular intervals. Despite this, we only included myopic subjects with at least 4 yearly follow-up visits over a period of 36 months and this may have restricted this limitation. Second, this is a retrospective study and future prospective longitudinal studies will be needed to confirm our findings. Another limitation is that we did not assess lacquer cracks among the potential risk factors for the development of recurrences. Lacquer cracks are thought to represent mechanical breaks of Bruch’s membrane,6 36 and their presence was demonstrated to be significantly associated with both myopic CNV development and recurrence rates.17 35 37 However, detection of lacquer cracks with conventional examination methods (eg, structural OCT) is difficult as they are better visualised using indocyanine green angiography,38 which was not available in the entire study population. Our study also has strengths, including the use of standardised data collected over 3 years of follow-up and the use of two masked, independent, and experienced graders for each image.

In conclusion, this longitudinal study provides imaging risk factors for development of recurrences in eyes with newly diagnosed myopic CNV that had resolved exudation after an initial treatment of anti-VEGF injections. These findings may facilitate the clinical care of patients with this disorder as clinicians may be assisted in informing patients of the risks and in delineating the best strategy for follow-up. This may eventually result in an overall improvement of visual outcome in these patients. Finally, our findings may be potentially relevant in the selection and randomisation of eyes with myopic CNV in future intervention clinical trials.

Data availability statement

Data are available upon reasonable request. Data are available upon request to the corresponding author.

Ethics statements

Patient consent for publication



  • Contributors Study concept and design: EB, MB, GV, FB and GQ. Acquisition, analysis or interpretation of data; critical revision of the manuscript for important intellectual content: all authors. Drafting of the manuscript and statistical analysis: EB. Study supervision: EB, FB and GQ.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests FB is a consultant for: Alcon (Fort Worth, Texas, USA), Alimera Sciences (Alpharetta, Georgia, USA), Allergan Inc (Irvine, California, USA), Farmila-Thea (Clermont-Ferrand, France), Bayer Shering-Pharma (Berlin, Germany), Bausch and Lomb (Rochester, New York, USA), Genentech (San Francisco, California, USA), Hoffmann-La-Roche (Basel, Switzerland), Novagali Pharma (Évry, France), Novartis (Basel, Switzerland), Sanofi-Aventis (Paris, France), Thrombogenics (Heverlee, Belgium), Zeiss (Dublin, USA). GQ is a consultant for: Alimera Sciences (Alpharetta, Georgia, USA), Allergan Inc (Irvine, California, USA), Amgen (Thousand Oaks, USA), Bayer Shering-Pharma (Berlin, Germany), Heidelberg (Germany), KBH (Chengdu, China), LEH Pharma (London, UK), Lumithera (Poulsbo; USA), Novartis (Basel, Switzerland), Sandoz (Berlin, Germany), Sifi (Catania, Italy), Sooft-Fidea (Abano, Italy), Zeiss (Dublin, USA).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Linked Articles

  • At a glance
    Frank Larkin