Ocular oncologists relate expert experience in identifying, treating retinal tumors in children.
In the past 15 years at one institution, no children with retinoblastoma at high risk for metastasis have died of the disease after early identification and treatment with chemotherapy following enucleation, according to an ocular oncology expert.
“Enucleation for retinoblastoma is not the end of treatment. It is just the beginning of treatment,” Carol L. Shields, MD, said of the Wills Eye Oncology Service/Children’s Hospital of Philadelphia experience.
At the American Association for Pediatric Ophthalmology and Strabismus annual meeting in San Antonio, where both she and husband Jerry A. Shields, MD, were honored with the Marshall M. Parks, MD, Silver Medal, Dr. Carol Shields told colleagues that, unlike melanoma, retinoblastoma is a known, predictable disease. Debatable, however, is how to define high-risk disease and what medications to use in those cases.
Carol L. Shields
Jerry A. Shields
Retinoblastoma is a highly malignant tumor that arises from the retinal cells of an embryo and develops in the first few years of life. Retinoblastoma affects 8,000 children worldwide, or 12 cases per 1 million children.
Death results when high-risk disease invading the orbit is not identified or treated, Dr. Carol Shields said. More than 3,000 children die each year worldwide of the disease, more so in developing countries than in the U.S., where only 4% of children die of the disease compared with 70% in Africa.
Typical white reflex in a child with retinoblastoma.
Image: Shields CL
According to Dr. Carol Shields, high-risk retinoblastoma is defined as invading areas where it is not supposed to be, for example, in the anterior chamber or postlaminar optic nerve, or massive invasion of the choroid greater than 3 mm.
When a patient has features of high-risk retinoblastoma, metastasis could occur, Dr. Carol Shields said; if there are no high-risk features, metastasis will not likely occur.
“Prelaminar invasion alone is not a risk,” she said. “Laminar invasion is not a risk. Postlaminar invasion is a risk for metastatic disease.”
Once retinoblastoma is identified as high risk, Dr. Carol Shields and colleagues implement a treatment protocol that incorporates a six-cycle regimen of chemotherapy with vincristine, etoposide and carboplatin.
“This is very effective therapy,” she said. “If you look at children with high-risk retinoblastoma and you don’t treat them with chemotherapy, 24% of them will be dead as a result of metastatic retinoblastoma.”
In a study published in Archives of Ophthalmology, Swathi Kaliki, MD, Dr. Carol Shields and colleagues reported outcomes of 55 patients with high-risk retinoblastoma who had undergone enucleation and then the chemotherapy protocol.
“In our study of 55 patients affected, 13 should have died,” Dr. Carol Shields said. None did.
A successful outcome for the high-risk retinoblastoma patient requires many people to work together, Dr. Carol Shields said, including the pathologist who identifies the histopathology and the ocular oncologist who interprets the findings along with the pediatric oncologist to start the chemotherapy.
“It is a collaborative effort,” she said.
“You must look at the pathology, and you must judge whether this child is at high or low risk for metastatic disease,” Dr. Carol Shields said.
Judgment based on experience, distinction of nomenclature, pathology, clinical exam and imaging, as well as new observations in cases of retinal pigment epithelium tumors, can allay a parent’s fears in some cases, according to Dr. Jerry Shields, who also spoke at the AAPOS meeting. Here, again, precise distinction of the disease process is key.
Parents worry when they hear the diagnosis of congenital hypertrophy of the retinal pigment epithelium (CHRPE) that their child will later develop colon cancer. However, based on new observations and 45 years of experience, Dr. Jerry Shields told colleagues unequivocally that two types of CHRPE have no association with either familial adenomatous polyposis (FAP) or Gardner syndrome, both hereditary syndromes.
“Solitary congenital hypertrophy of RPE and multifocal CHRPE have no relationship to bodily cancer,” he said.
There is, however, another type of CHRPE that is nearly 100% associated with development of life-threatening colon cancer: FAP-associated CHRPE, which is characterized by multiple bilateral lesions.
When the term CHRPE is used indiscriminately to identify any one of the three types of CHRPE, parents conclude that development of FAP is inevitable, but there are distinct differences among the three types of CHRPE.
CHRPE is characteristically a unilateral, deeply pigmented, flat lesion with a sharply demarcated margin. Optical coherence tomography shows an absence of photoreceptors corresponding distinctly to the lesion, which goes along the absolute scotoma that is often seen in these patients, Dr. Jerry Shields said. The lesion is usually uniformly black with a well-delineated marginal halo and depigmented lacunae.
Of the solitary type of CHRPE, Dr. Jerry Shields said, “It has been believed for years that this is a solitary lesion that will not change.”
However, over the years, Dr. Carol Shields, Dr. Jerry Shields and colleagues have obtained serial studies on hundreds of patients and found that solitary CHRPE lesions can increase in diameter in up to 80% of cases. Furthermore, through their studies, they found that solitary CHRPE lesions can spawn nodular tumors in 2% of cases. In a few cases, solitary CHRPE evolved into malignant epithelioma, producing severe complications, but there is no known metastatic potential, he said.
Dr. Jerry Shields advised yearly follow-up for these patients. – by Sara E. Olson and OSN Retina Staff
- Kaliki S, Shields CL, Shah S, Eagle RC Jr, Shields JA, Leahey A. Postenucleation adjuvant chemotherapy with vincristine, etoposide, and carboplatin of high-risk retinoblastoma. Arch Ophthalmol. 2011;129(11):1422-1427.
- Kivelä T. The epidemiological challenge of the most frequent eye cancer: retinoblastoma, an issue of birth and death. Br J Ophthalmol. 2009;43(9):1129-1131.
- Shields CL, Mashayekhi A, Ho T, Cater J, Shields JA. Solitary congenital hypertrophy of the retinal pigment epithelium: clinical features and frequency of enlargement in 330 patients. Am J Ophthalmol. 2003;110(10):1968-1976.
- Shields JA, Eagle RC Jr, Shields JA, Brown GC, Lally SE. Malignant transformation of congenital hypertrophy of the retinal pigment epithelium. Ophthalmology. 2009;116:2213-2216.
- Shields, JA, Shields CL. Tumors and related lesions of the retinal pigment epithelium. Intraocular Tumors: An Atlas and Textbook. 2nd ed. Philadelphia: Lippincott, Williams and Wilkins; 2008:432-463.
- Shields JA, Shields CL, Shah PG, Pastore DJ, Imperiale SM Jr. Lack of association among typical congenital hypertrophy of the retinal pigment epithelium, adenomatous polyposis, and Gardner syndrome. Ophthalmology. 1992;99(11):1709-1713.
For more information:
- Carol L. Shields, MD, is associate director of Ocular Oncology Service and can be reached at Wills Eye Institute, 840 Walnut St., Philadelphia, PA 19107; 215-928-3105; email: firstname.lastname@example.org.
- Jerry A. Shields, MD, is director of Ocular Oncology Service and can be reached at Wills Eye Institute, 840 Walnut St., Philadelphia, PA 19107; 215-928-3105; email: email@example.com.
- Disclosures: Drs. Carol and Jerry Shields have no relevant financial disclosures.