• Why we use cookies

    Our website uses 1 cookie. A cookie is a small file of letters and numbers that we put on your device if you agree. These cookies allow us to anonymously distinguish you from other users of our website, which helps us to provide you with a good experience when you browse our website and also allows us to improve our site. The cookies we use are analytical cookies. They allow us to recognise and count the number of visitors and to see how visitors move around the site when they are using it. This helps us to improve the way our website works, for example by ensuring that users are finding what they are looking for easily. By accessing our website you are deemed to have consented to the setting of the cookie on your device.

Previous CHECT projects

Here are the details of research projects previously funded by the Childhood Eye Cancer Trust:


Epigenetic Mechanisms in Retinoblastoma Tumourigenesis

Project Leader: Dr Carmel McConville at the CRUK Institute for Cancer Studies at the University of Birmingham.

Funding awarded: £50,500

Project duration , Start date: 1 year from February 2012

Project summary (As written by Dr C McConville) : Research into new treatments for all types of cancer is increasingly focusing on the development of 'targeted therapy'. There is a need to develop new drugs which are very specific in their action and which target cancer genes that are known to be defective in particular types of cancer. In the case of retinoblastoma it is known that a gene called 'DEK' is present at abnormally high levels in approximately 50% of cases. It is possible that DEK might contribute to the growth of the retinoblastoma and perhaps also to its spread outside the eye. It could therefore be a target for therapy in these retinoblastomas.

The purpose of our research is firstly to find out what effect DEK has on retinoblastoma cells. Does it make them grow faster? Does it make them more mobile and invasive? Or does it influence their ability to carry out the specialized functions of the retina? We will also try to find out exactly how DEK works. Current evidence suggests that the protein encoded by the DEK gene works in association with other proteins to 'switch on' or 'switch off' other genes. This might include genes which are important for normal retinal development and function. Understanding exactly how DEK does this will be the key to developing new drugs to block its action.

Our research will involve growing retinoblastoma cells in the laboratory. We will use a specialized genetic technique ('gene knockdown') to silence the DEK gene and we will then monitor the changes which occur. We will also investigate the interactions between DEK and other proteins in order to understand the mechanisms involved.

Project status: Project completed, final report here. Awaiting publication of results.

Back to top

Epidemiological studies of retinoblastoma with particular reference to the incidence of multiple primary tumours.

Project leader: Dr Gerald Draper, Childhood Cancer Research Group, Oxford
Funding award: £34,879 in collaboration with University of Oxford, Childhood Cancer Research Group
Project duration: April 2003 – March 2006 Three years
Introduction: The Childhood Cancer Research Group (CCRG) has records of virtually every case of retinoblastoma (RB) occurring in Great Britain during the past forty years, together with records of selected groups of cases in earlier years. This database is linkable to laboratory records relating to many of these cases, including, in particular, those held at the Retinoblastoma Screening Unit at St Bartholomew's Hospital (Dr Zerrin Onadim). In addition, the group has access to pedigree data collected by other research workers and to international data on the incidence of retinoblastoma. It is proposed, within the CCRG's ongoing programme of epidemiological research, to update earlier analyses of these data and to initiate others.
1. Multiple primary tumours
In particular, we propose first to carry out a comprehensive analysis of the incidence of multiple primary tumours in RB survivors in Britain. This would be followed by a systematic review (and possibly a meta-analysis) of other published and, if possible, unpublished, data so as to provide the best possible estimates of the rate of occurrence of other cancers following survival from RB, and a description of the types of cancer occurring. In addition, at least for the British data, we shall attempt to relate the risk of other cancers to the results of mutational and chromosomal analyses.
2. Risks to offspring and siblings.
The second part of the project will be the analysis of a new dataset on the incidence of RB in the offspring of surviving RB patients. These results will then be combined with those from other studies in order to provide updated estimates of risks to offspring of RB patients. Similar analyses will be carried out for siblings.
3. Incidence rates of retinoblastoma (Rb).
The third part of the project will be a systematic review of both published and unpublished data concerning international variations in the incidence of retinoblastoma (Rb), with particular reference to separate analyses of unilateral and bilateral cases in the hope that such analyses will help to elucidate the aetiology of this condition and to estimate its public health importance in different countries.
Project Status:
Two of the three stages complete, official publication of the Second Primary Cancer research is pending.

Final Report

The final lay report can be viewed here.

Retinoblastoma in Great Britain 1963-2002
Angela MacCarthy, J M Birch, G J Draper, J L Hungerford, J E Kingston, M E Kroll, Z Onadim, C A Stiller, T J Vincent and M Murphy
Br. J. Ophthalmol.published online 6 Oct 2008; doi:10.1136/bjo.2008.139618

Retinoblastoma: Treatment and Survival in Great Britain 1963 to 2002
Angela MacCarthy, J M Birch, G J Draper, J L Hungerford, J E Kingston, M E Kroll, C Stiller, T Vincent and M Murphy
Br. J. Ophthalmol. published online 6 Oct 2008;

Non-ocular tumours following retinoblastoma in Great Britain 1951 to 2004
A MacCarthy,1 A M Bayne,1 G J Draper,1 E M Eatock,1 M E Kroll,1 C A Stiller,1
T J Vincent,1 M M Hawkins,2 H C Jenkinson,3 J E Kingston,4 R Neale,5 M F G Murphy1
doi: 10.1136/bjo.2008.146035
Br J Ophthalmol 2009 93: 1159-1162

Back to top

The role of HMGA1 and HMGA2 in retinoblastoma

Project duration: 6 months, 2005
Grant awarded: £4,817.33
Project lead:
Mr A M Reddy
Introduction: Retinoblastoma is the commonest primary intraocular tumour in children. Chemotherapy is an important and essential modality in the treatment of retinoblastoma. High Mobility Group A2 protein (HMGA2), a nuclear protein, which is widely expressed in embryonic tissue and down regulated in nearly all adult tissue. HMGA2 expression is known to occur in other tumours such as liposarcomas, a secondary non-ocular tumour associated with retinoblastoma. Research from Prof Ono's laboratory at the Institute of Ophthalmology, London has shown that 3 established RB cell lines and 3 of 4 primary tumour samples express HMGA2. Interestingly, anti-sense gene therapy, which reduces the expression of this protein, reduced RB cell proliferation in vitro.
Aim: To investigate the expression of HMGA1 and HMGA2 proteins in archival retinoblastoma tumors
Methods: Sixty archival tumour samples will be obtained from the ocular pathology Lab of Vision Research Foundation, Sankara Nethralaya. The RB tumours will be divided into 2 groups. Group A: RB with no invasion and Group B: RB with invasion of choroids, optic nerve. HMGA2 protein expression will be analysed using immunohistochemistry on the 2 groups of tumours. Five fresh tumour samples will be also subjected to western blotting to confirm the immunohistochemical data. The expression of this protein will be analysed in a semiquantitaive manner. Information on the chemotherapy will be also looked into. Normal ocular tissues expressing HMGA2 will be looked for.
Outcome: The study will give us information whether HMGA2 is expressed uniformly throughout the RB tumours or there is a heterogeneous expression in the tumour samples. Also HMGA2 expression will be correlated with invasion of the tumours. The study will also generate data whether the HMGA1 protein is expressed in non-neoplastic retinal tissues. This has implications in damage to the non-neoplastic retinal tissues.

Project status: Complete

The results from this project led to the funding of a subsequent project entitled 'To study the RNAi mediated HMGA2 gene expression profile of y79 cells'. The details of this project are also on this page.

Back to top

Psychological consequences of treatment for Rb in young children

Project duration: 1 year, October 2002
Grant awarded:
Project lead:
Dr S Norgate
Introduction: The onset of retinoblastoma early in a child's life puts at risk aspects of the child's development and family wellbeing. Crucially, there is substantial deterioration in vision during the precise developmental period when sighted peers are using sight to learn to talk, move round their surroundings and understand aspects of their social and physical environment. Further, the diagnosis, the associated treatment process, as well as the loss of vision has the potential to severely compromise the mental health of both the child and the whole family over many years. The present proposal seeks to examine the psychological consequences of retinoblastoma and associated treatment in the under-fives.
Aims: Through talking with members of the Retinoblastoma Society, clinical staff and families we have devised three research objectives.
The first objective is to ascertain how bilaterally affected children co-ordinate their remaining vision, sound and touch after treatment.
The second objective is to develop an age-related profile of the understanding that unilateral, bilateral and age-matched sighted children have of eyes.
The third objective arises from the recognition of the range of factors affecting children with retinoblastoma and their families.
Method: It is important to talk, in in-depth interviews, with consenting families about their experiences and wishes in relation to aspects of the treatment process and their child's development. In the context of the individual needs of each family and the financial and time constraints of service delivery, the aim is to identify those aspects of the situation and experience which are changeable.
Outcome: We intend to use the findings from each of the objectives to identify and promote coping strategies and also to identify aspects of development at risk. Throughout the project we would hope to work in tandem with clinical staff and seek advice concerning how to communicate outcomes and feedback to families.

Final report

The final report is available here.

Back to top

1. Is tumour size predictive of response to chemotherapy? 2. Does laser hyperthermia of retinoblastoma foci increase the risk of vitreous seeding?

Project duration: 10 months, January 2000
Grant awarded:
Project lead:
Mr D S Gombos
Project 1
Introduction: Over the past decade primary chemotherapy has been increasingly used to treat multi-focal bilateral retinoblastoma. This approach has the potential advantage of avoiding external beam radiotherapy, which can alter the development of orbital tissues and predispose children to second cancers. The response to chemotherapy however varies among patients. In some cases tumors dramatically decrease in size while in others little change is seen. New tumors have also been observed developing in some patients concurrently receiving chemotherapy. Further research is necessary to identify the factors that correlate with better tumor response. This would allow chemotherapy to be used more effectively; patients expected to have a poor response could be offered adjuvant or alternative therapy (such as laser hyperthermia, cryotherapy or plaque radiotherapy) earlier in their course of treatment.
We have observed a trend suggesting that the size of a tumor at diagnosis may correlate with its response to chemotherapy. Larger tumors seem to have a large regression following treatment while the response of smaller tumors is more variable.
If the response of retinoblastoma foci to chemotherapy is related in part to their blood supply, one would expect bigger tumors to be more responsive. Large tumors are often observed having 'feeder vessels' supplying them with blood. Conversely smaller tumors, with less blood supply may not receive appropriate concentrations of chemotherapy necessary to arrest their growth.
Aims: The purpose of this study is to evaluate whether the size of a tumor at diagnosis correlates with its response to chemotherapy. Is the size of a tumour at diagnosis predictive of its response to chemotherapy alone?
Methods: A retrospective chart review will be performed on all patients diagnosed at St. Bartholomew's Hospital with retinoblastoma and treated with chemotherapy. Those patients who received adjuvant therapy such as laser hyperthermia or radiotherapy will be excluded. The basal size of each tumor at diagnosis will be obtained from a combination of retinal drawings, ultrasound measurements and ret-cam fundus photography. Each tumor will them be reviewed with regards to its response to treatment. Statistical methods will be employed to evaluate the data and identify any positive or negative correlations between the size of tumor foci and their success or failure of treatment with chemotherapy.
Project 2
Introduction: Untreated, retinoblastoma carries a near 100% fatal prognosis. The successful use of radiotherapy, chemotherapy and cryotherapy has significantly altered not only patient mortality but ocular morbidity as well. One or both eyes are often preserved with excellent visual results. Increasingly, chemotherapy and adjuvant laser hyperthermia have become the primary agents used to treat this disease.Among the few conditions where the prognosis for ocular salvage remains poor is vitreous seeding. In this advanced stage, tumor cells invade the central jelly of the eye (the vitreous).
These eyes respond poorly to chemo- and radioth radiotherapy and often succumb to enucleation. Relapse of tumor into the vitreous is a rare finding but recently a greater number of these cases have been observed on our service. Some patients developed vitreous seeding shortly after treatment with laser hyperthermia.
Due to the difficulty in treating eyes with vitreous seeding it is important to establish if eyes treated with laser hyperthermia carry a greater risk of developing vitreous relapse. If so, are there any factors (such as certain laser settings) that correlate with this increased risk.
Aims: The purpose of this study is to establish whether laser hyperthermia correlates with an increased risk of vitreous seeding.
Methods: A retrospective chart review will be performed on all patients diagnosed and treated with retinoblastoma at St. Bartholomew's hospital over the past 5-6 years. Individual tumor foci treated with laser hyperthermia will be reviewed and data collected regarding laser settings used. Rates of vitreous relapse will then be calculated among eyes treated with and without laser hyperthermia. A statistical analysis will be employed to identify any correlations between particular laser settings and rates of vitreous disease.

Final report

Retinoblastoma treated with primary chemotherapy alone: The significance of tumour size, location and age.

Dan S Gombos MD 1,2; Alison Kelly 1; Pietro G Coen D Phil 4; Judith E Kingston FRCP 1,3; John L Hungerford FRCS 1,2


Background/aims: To evaluate how tumour size, retinal location and patient age affect the outcome of retinoblastoma foci treated with chemotherapy.
Methods: Retrospective review of retinoblastoma foci treated with primary chemotherapy alone. Individual tumours were coded with regard to their largest basal diameter, location within the eye (macula, macula to equator, equator to ora serrata) and patient's age at diagnosis. Successfully treated tumours required no further intervention while those requiring additional treatment were coded as failures.
Results: Fifty-six (72%) tumours responded successfully to chemotherapy alone while 22 (28%) required additional therapy. Twenty-six of 31 macular tumours (84%) and 30 of 47 extramacular tumours (64%) responded to chemotherapy (p < 0.060). Relative to size, 46 of 60 tumours (77%) greater than 2 mm in basal diameter were successfully treated with chemotherapy, while only ten of 18 tumours (56%) less than or equal to 2 mm responded (p < 0.088). Among the eight tumour foci diagnosed in children less than two months of age, five (63%) failed to respond to chemotherapy alone (p < 0.032).
Conclusion: Retinoblastoma is more likely to respond to primary chemotherapy if it is located in the macula and if the patient is greater than two months of age. Tumours measuring less than 2mm in diameter may be less responsive to this modality.

Back to top

Childhood Cancer Research Group: Development of retinoblastoma database.

Project duration: May 1997, 1 year
Grant awarded: £19,528
Project lead: Prof. G Draper & Dr J Kingston

Retinoblastoma database project final report

The database project was started in May 1997 with a project grant from the Retinoblastoma Society. The aim of the project was to create a retinoblastoma database to facilitate research on the clinical, molecular and epidemiological data that was accumulated over the years. A retinoblastoma database based on Microsoft Access software was created by Pat Brownbill. This database was designed for molecular and clinical data entry and for routine use in the laboratory by the Retinoblastoma Genetic Screening Unit (RGSU). Retinoblastoma patient details, molecular data and some clinical data dating from October 1988 (covering Zerrin Onadim's data from ICH and all of RGSU data) was entered into this database by Michele Stagg working part-time between May 1997 to April 2000. After this date the Retinoblastoma Unit continued to enter laboratory data into database to keep it up to date.
In the laboratory, this database is used regularly for registration of new patients and addition of details of existing patients.  Diagnosis, treatment, surgery, cytogenetics tests details, every blood and tumour samples received, all DNA/RNA samples stored from each patient are entered.  Details of all gel runs are also recorded.  The test results in the form of polymorphism data and mutations identified are entered.  The database has a separate enzyme and RB1 gene sequence section and enzyme site search facility.  The database can be used to perform various searches on patient data including gels run and mutations.  There is also an administration section containing addresses and contact details of professionals.
To date, data for 2600 individuals were entered.  1077 of these entries are from affected patients.  Among these are 559 cases of bilateral, 480 cases of unilateral and 4 cases of trilateral Rb.  There are 128 different Rb families (i.e. families in which there are at least two cases of Rb -familial Rb-), 28 unaffected mutant RB1 gene carriers and 14 regressed Rb cases on record.  The number of mutations entered into the database to date is 253.
The database is already an indispensable tool for clinical and laboratory work.  As part of the project 'Epidemiological studies of retinoblastoma...' being led from the Childhood Cancer Research Group, the mutation data will be linked to data on multiple tumours and family histories held by that group so as to try to identify possible relationships between type of mutation and (a) the risks of subsequent non-ocular cancers occurring in Rb patients (b) the risks of Rb in their siblings and children.

Back to top

Retinoblastoma RB1 gene mutation analysis: Molecular Genetic Study of Hereditary Retinoblastoma

Project duration: December 1997, three years
Grant awarded: £91,989 in collaboration with Queen Mary & Westfield College
Project lead: Dr. Z Onadim


This thesis describes a molecular genetic analysis of retinoblastoma (Rb), the commonest ophthalmic malignancy in childhood, that has both familial and non-familial forms. In familial cases the disease is transmitted as an autosomal dominant trait with incomplete penetrance. At the start of the study the gene responsible, RB1, which was also the first tumour suppressor gene identified, was available together with several intragenic probes.
The aim of the study was to analyse mutations in patients predisposed to Rb to gain an insight into the mechanism of mutagenesis and to offer screening for carrier status. The first part of the study involved family linkage analysis with Restriction Fragment Length Polymorphisms (RFLP) and other intragenic polymorphisms of the RB1 gene using Southern blotting techniques and the Polymerase Chain Reaction (PCR). The results indicated that, using the combination of available polymorphisms, it is possible to offer screening to 95% of Rb families. Using PCR, it was also possible to use formalin-fixed, paraffin-embedded tissue samples from archival material. During the course of this analysis, five pre-natal and six post-natal screenings were carried out the results of which indicated inheritance of the predisposing allele in three cases.
In search of a suitable technique to identify causative mutations in the RB1 gene various techniques were assessed for their usefulness. These techniques were Ribonuclease (RNase) protection, PCR sequencing of individual exons, Carbodiimide (CDI) modified heteroduplex analysis and Single Strand Conformation Polymorphism (SSCP) analysis. SSCP analysis proved to be the most successful in this study and six germ-line mutations were identified in the RB1 gene. Four of the mutations were from typical Rb families with bilaterally affected individuals and the other two from families with “mild” phenotypes. The analyses of these mutations provided insights into the mechanisms of mutagenesis in RB1 and the phenotype-genotype relationship in Rb.

Final report

The final report is available to read here.

Back to top

Identification of molecular tools.

Project Leader: Dr Carmel McConville at the CRUK Institute for Cancer Studies at the University of Birmingham.
Funding awarded: £70,000
Project duration:3 years.
Project Summary (As written by Dr C McConville) Survival rates for retinoblastoma in the UK are well in excess of 90% and the challenge in treating this disease is to maximize preservation of sight whilst minimizing the risk of recurrent disease and treatment-related toxicity. The purpose of this project is to provide a molecular characterisation of retinoblastoma which will allow a more informed approach to balancing these twin objectives. We will investigate differential gene expression in retinoblastomas with and without adverse pathological features, in order to identify genetic events associated with tumour progression. We will also investigate the potential of magnetic resonance spectroscopy (MRS) to provide a non-invasive 'molecular biopsy' of the tumour at the time of diagnosis.
We anticipate that this work will allow identification of molecular markers which may be used to predict tumour behaviour and to guide treatment decisions. The identification of aberrantly regulated genes/genetic pathways will also provide an opportunity for the development of targeted therapy, which could potentially reduce the requirement for enucleation.
Project status:Completed
Final Report (As written by Dr C McConville):Finding the best treatment for retinoblastoma presents some unique challenges. The first challenge is not just to cure the cancer but also to try to preserve sight - avoiding enucleation if possible. A second challenge is the development of a 'personalised medicine' approach to therapy.
Not all retinoblastomas are the same and selection of the optimal treatment for each patient should take into account not just the clinical and pathological features of the cancer but also, perhaps more importantly, the genetic characteristics.
Mutation of the RB1 gene is necessary, but not sufficient for the development of retinoblastoma in isolation. Relatively little is known about other genetic abnormalities which contribute to development of the cancer, and which also determine how it will behave (e.g. propensity for optic nerve invasion). The aim of our CHECT research project was to investigate high grade retinoblastomas (i.e. those which are often enucleated) from 20 different patients in order to determine the extent of genetic variability among these retinoblastomas, to identify the genetic pathways involved and to determine if these are correlated with clinical characteristics.
Our results showed very clearly that not all retinoblastomas clinically classified as high grade (Group D or E), are the same at the genetic level. Most retinoblastomas fell into one of two main groups (Figure 1). Group 2 retinoblastomas had some genetic similarities to cone photoreceptor cells (for a description of retinal cell types see ). In contrast group 1 retinoblastomas did not clearly resemble any specific retinal cell type but may be derived from a progenitor-type cell (slightly more progressed than a stem cell) which has not yet developed a specific identity or function in the retina. Significantly, the frequency of optic nerve or choroid invasion was more frequent in patients with Group 1 retinoblastomas suggesting that these may represent a more aggressive cancer type. Group 2 retinoblastomas on the other hand may be less aggressive and patients with these retinoblastoma may benefit from more conservative treatment.
This research is a first step towards the development of customised treatment for retinoblastoma patients and we are currently working further towards this goal, through analysis of the genetics of a much larger number of patients, and by investigation of genes specific to Group 1 retinoblastomas, which could potentially be targeted using novel therapeutic agents.

Back to top

To study the RNAi mediated HMGA2 gene expression profile of y79 cells

Project Leader: Prof Krishnakumar from the Sankara Nethralaya Centre in India and Mr Ashwin Reddy, Consultant Paediatric Ophthalmologist with interest in Retinoblastoma at Royal London Hospital.
Funding awarded:£15,000
Project duration:3 years.
Project Summary: Understanding proteins that produce invasive retinoblastoma: The role of HMGA2 (High Mobility Group A2) in the development of retinoblastoma, summary by Mr Ashwin Reddy
We still do not understand why some retinoblastoma tumours grow slowly and some are very aggressive. The more aggressive retinoblastoma tumours (invasive retinoblastoma) invade into the outer layers of the eye. We are aware that loss of the retinoblastoma protein initiates events with tumour development, but other proteins are also implicated.
Dr Krishnakumar’s team in Chennai, India, working with Mr M. Ashwin Reddy, Retinoblastoma surgeon at the Royal London Hospital, has already shown that a protein called HMGA2 (High Mobility Group A2 protein) is produced more in tumours that are invasive.
Another paper that has recently been accepted and funded by CHECT gives an explanation as to why. When this protein is not produced, retinoblastoma cells stop growing. Therefore this protein appears to be very important for the growth of the tumour-forming cells.
The genes that make proteins for tumour effectiveness such as ‘stickiness’ to blood vessels are affected. A tumour with more ‘sticky’ proteins is likely to be more aggressive. Other genes that are important for tumours to work are also affected.
This is more evidence that stopping HMGA2 from working may be beneficial for patients as the retinoblastoma tumours may stop growing. It also shows how complex the biological pathways are for retinoblastoma.

Project status: Complete

Full free paper

AMRC  logo

CHECT projects under way

Find out which new projects have been funded by us

How to apply for funding

Apply for a research grant

Scientific Advisory Committee

Find out who sits on our Scientific Advisory Committee

Back to top