The Evolution of Treatments for Retinoblastoma
After decades of advances, superselective chemotherapy alone can now destroy tumors and result in useful vision in the majority of patients.
The management of intraocular retinoblastomahas changed dramatically over the past century,and as a result, so have the outcomes in terms ofsaving the lives, eyes, and vision of our patients.Retinoblastoma was almost universally fatal a little morethan 100 years ago, but now in the Western world it hasthe highest cure rate of any pediatric solid tumor.1
How did we get from there to here? Treatment algorithmsfor retinoblastoma advanced periodically duringthe 20th century, and as a result, outcomes in terms ofsurvival and the preservation of eyes and vision steadilyimproved. Now in the first decade of the 21st century, anemerging treatment paradigm promises to further revolutionizethe management of the disease worldwide. Thisarticle reviews the path to where we are today and thepotential of this new treatment approach to preserve ourpatients’ vision and improve their quality of life.
TOWARD SIGHT PRESERVATION
In the late 1800s and early 1900s, there were few survivorsof retinoblastoma. The only treatment was enucleation,and usually the tumor was diagnosed at a stage atwhich removing the eye did not save the patient’s life.With enucleation, the survival rate was perhaps 5% at theturn of the 20th century.
Successful treatment of retinoblastoma with radiationwas first reported in 1903.2 (X-rays were used therapeuticallyfor cancer treatment within a year of their discoveryby Roentgen in 1895, and for retinoblastoma only a fewyears later.) From that time until the 1940s, the use ofradiation became more and more common.
And it cured the cancer; retinoblastoma was, and still is,one of only a few malignancies in children that can becured by radiation alone. But although more patients nowsurvived as techniques and dosages for radiation wererefined, and some kept their eyes, none retained vision.
In England in the 1930s, Stallard3 introduced radioactiveplaques that allowed radiation to be delivered toonly a portion of the eye without irradiating the whole eye and orbit. In the 1940s, Reese and colleagues4 in NewYork devised methods to deliver external radiation thatsaved eyes and preserved vision. These developmentswere accompanied by a dramatic change in treatmentphilosophy. The standard protocol until that that timewas to remove the worse eye surgically and treat the lessinvolved eye with radiation. Now, more patients survivedand kept their eyes, and in some cases, their vision.
In the 1950s, Meyer-Schwickerath5 in Germany introducedthe concept of photocoagulation with white lightto destroy small tumors. It was soon recognized thatoccasionally retinoblastoma could be treated with photocoagulationalone, without enucleation or radiation,and that this could preserve eyes with some vision.
In the 1960s, Lincoff6 introduced the use of cryotherapy,which, like photocoagulation, was successful in destroyingsmall tumors and preserving vision in some eyes.
Until the late 1980s, these were the tools available fortreatment of retinoblastoma, and with these tools, bythat time, the survival rate for retinoblastoma in theWestern world exceeded 90%. (It is a sad but well documentedfact that, worldwide, the majority of childrenwith retinoblastoma still die, even in 2010.)
As survival improved in patients treated with radiation,however, clinicians began to recognize that in the longterm, these patients often developed second nonocularcancers in the irradiated field.7 In addition, as genetictesting became possible, it was noted that second (andthird, and fourth) tumors were more common inpatients with the RB1 gene mutation.8
Because these secondary cancers occurred at a rate of0.5 to 1% per year, and because half of the children whodeveloped these other cancers died, by the 1990s, themost common cause of death in retinoblastoma patientswas not the retinoblastoma itself but secondary cancersrelated to the patients’ genetics or to their radiationtreatment.8
As clinicians began to look for alternatives toradiation —not because it did not work butbecause of these long-term complications—the use of systemic chemotherapy was widelyadopted. Systemic chemotherapy forretinoblastoma was first described by Kupfer9in the 1950s, but interest grew starting in the1990s, and there are now more than 150 publicationson the subject. These publications haveshown that three cycles of carboplatin-basedchemotherapy over 3 months can reduce thesize of tumors by almost 50%.10 Unfortunately,chemotherapy alone rarely cured the tumors,but if they were reduced to a small enoughsize, they could be treated with laser, cryo, plaques, or externalradiation. With these adjunctive treatments, the successrate of systemic chemotherapy was comparable to radiationbut without the radiation-related side effects.
So the good news about systemic chemotherapy wasthat it worked against retinoblastoma, and it has workedin many investigators’ hands—leading to the aforementioned150-plus publications on the subject. The badnews was that, alone, it did not cure cancer, so nearly allpatients needed additional treatment with radiation,plaque, laser, cryo, or even enucleation. This greatlyincreased the amount of time necessary to treat and curepatients, and the costs, both economic and in increasedburdens, on families.
But there are other problems with systemic chemotherapyfor retinoblastoma as well, which have resulted inwidespread disenchantment with this approach. Oneside effect of chemotherapy in these children has beenhearing loss, which is, of course, unfortunate in childrenwho already have vision problems; rates of 5 to 33% permanenthearing loss after chemotherapy have beenreported.11,12 In one report from Mexico, where cisplatinwas used, 100% of children experienced permanent hearingloss.13
In addition, common complications of chemotherapyinclude the need for transfusions and ports and theoccurrence of febrile neutropenia. Toxicities, includingsecond cancers, can result from the chemotherapy aswell. Multiple studies have found that children receivingchemotherapy and radiation are at higher risk for developmentof second cancers than those receiving radiationalone.14-17 Secondary acute myelocytic anemia, a virulentform of leukemia that is difficult to treat and has a highmortality rate, has been reported after chemotherapy in15 children with retinoblastoma18 and is now the mostcommon second cancer in some countries.
Because of these issues with systemic chemotherapyfor retinoblastoma, clinicians have sought more selective(less toxic) ways of treating it. At Memorial Sloan-KetteringCancer Center in New York several years ago, we began toinvestigate a technique we call superselective ophthalmicartery chemotherapy.19-22 Beginning in 2006 as an institutionalreview board-approved experimental protocol, thetechnique was so successful that it has now replaced systemicchemotherapy at our center for all children over theage of 3 months. It has completely replaced radiation; wehave not used primary radiation in any child in the past 4.5years. This intraarterial mode of treatment is more effective,faster, easier on child and family, less expensive, and lesstoxic than any other treatment modality available. It hasnow been taught and performed in 20 countries worldwide.
In the intraarterial technique, a catheter is inserted intothe femoral artery on an outpatient basis with the childunder general anesthesia, passed through the abdominaland thoracic aorta into the internal carotid artery, and thenplaced directly in the ophthalmic artery, the only blood vesselsupplying the eye. A tiny volume of a chemotherapeuticagent or agents—between 0.5 and 1 cc depending on thesize of the child—is infused over the course of 30 minutes.
Our most recent publication22 reported our 3-year experiencewith 23 newly diagnosed patients with retinoblastomatreated with 75 separate intraarterial chemotherapyinfusions. Cannulation of one or both ophthalmic arteries(five patients had bilateral disease) was achieved in all eyes(Figure 1). All children survived, and only one proceeded toenucleation for progression of disease. No patient requiredtransfusion or hospitalization, no febrile neutropenia wasseen, and ocular complications included only transientforehead hyperemia, lid edema, and loss of lashes.
We have now performed this procedure more than300 times. For the first time, primary chemotherapy alone has cured a solid cancer in children (except forgerm cell tumors and choriocarcinoma); 40% of our childrenhave received only superselective chemotherapy,and 90% have retained the eye, many with useful vision.
Superselective ophthalmic artery delivery is a way ofincreasing the chemotherapy dose to the tumor anddecreasing it to the body. While we do not know theexact intraocular concentration of the chemotherapydrug, we know it is in excess of 100 times the concentrationthat would kill a human if administered at thesame concentration systemically. But these children donot get sick because the dose to the body is so small;they do not lose their hair or their appetites. Extensiveevaluation has shown that the treatment is not toxic tothe eye if the eye has not previously been treated withsystemic chemotherapy and/or radiation. In fact, insome children who have no vision before infusion, theretina settles down and vision is achieved.
This treatment works in the eye because at the timeof diagnosis, retinoblastoma is localized to the eye. Therest of the body does not need chemotherapy, whichunfortunately is not true in many cancers. It also worksin the eye because the doses and the drugs we use areeffective and not toxic. Retinoblastoma is very sensitiveto chemo, but in the conventional doses used systemically,it is not curative.
As a result of our work, a number of centers, includingMemorial Sloan-Kettering, are now investigating theuse of this approach for other cancers. Memorialrecently opened a new center dedicated to thisapproach for cancer.
The proof of a technique lies in its replication byother centers, and superselective ophthalmic arterychemotherapy is now being done successfully in20 countries. Not only have our results been replicated;in some cases, reports suggest even better results.Notably, many of these countries are emerging or developingnations. As noted above, although more than90% of patients with retinoblastoma survive in theWestern world, the rate of survival worldwide is lessthan 50%. It is encouraging that this high-tech procedure,on the cutting edge of treatments for retinoblastomaand potentially other types of cancers, is succeedingin the hands of clinicians in the developing world.I predict that by the end of 2010 or early 2011 this procedurewill have been performed in more emergingcountries than developed countries.
It has been quite a journey from the turn of the 20th centuryto the turn of the 21st. With conventional treatments,retinoblastoma has the highest cure rate of any childhoodcancer, and we should all be proud of the contributions ofresearchers and clinicians over the past 100 years to achievethat. At the same time, that achievement makes it a challengeto introduce further improvement.
Experience to date suggests that superselective ophthalmicartery chemotherapy is more effective, faster,better, and safer than conventional treatments for thiscancer, which already had good success rates. Now wecan address not only the survival of our patients, butalso the quality of their lives, in a way that matters. Likethe Wright brothers at the dawn of aviation, I look forwardto the future development of this new, transformationalapproach to cancer.
David H. Abramson, MD, FACS, is Chief of theOphthalmic Oncology Service at MemorialSloan-Kettering Cancer Center in New York City.He can be reached at +1 212 639 7232; or viae-mail at Abramsod@MSKCC.ORG.