HEALTH PROFESSIONAL · SOURCE READING
Subsequent Neoplasms (SNs)
Source: Retinoblastoma Treatment (PDQ®)–Health Professional Version, National Cancer Institute.
Source updated: April 3, 2025 · Captured 2026-09-09.
Selected source text with whitespace normalised. This Triangle page is not an NCI PDQ summary.
Context: Late Effects of Retinoblastoma Therapy
SNs are the most common cause of death in patients with retinoblastoma. SNs contribute to about 50% of deaths in patients with both bilateral disease and genetically defined heritable retinoblastoma.[2-4] Survivors of retinoblastoma have a high risk of developing SNs.
Factors that influence the risk of SNs include the following:
Heritable retinoblastoma. Patients with heritable retinoblastoma have a markedly increased incidence of SNs, independent of treatment with radiation therapy.[2,3,5] In a German series of 643 patients with heritable retinoblastoma, chemotherapy with or without radiation therapy was the only significant risk factor for the onset of second cancers outside the periorbital region.[3] A possible association between the type of RB1 variant and incidence of SNs may exist, with complete loss of RB1 activity associated with a higher incidence of SNs.[6] Patients who are heterozygous for regular-penetrance RB1 variants may be at a higher risk of developing SNs than patients with incomplete-penetrance RB1 variants.[7] Regular penetrance refers to patients with heritable disease and classic multifocal bilateral tumors. Incomplete penetrance, while rare, is a clinical phenotype in which affected family members may not develop retinoblastoma or they present with unilateral disease. With the increase in survival of patients with heritable retinoblastoma, it has become apparent that they are also at risk of developing epithelial cancers late in adulthood. A marked increase in mortality from lung, bladder, and other epithelial cancers has been described.[8,9]In a large series from two institutions, 2,053 patients with retinoblastoma (diagnosed between 1914–2016) were identified, with a maximum of 70 years of follow-up. Most deaths occurred in patients with hereditary retinoblastoma (518 of 1,129), and 267 of these deaths were caused by SNs. Increased risk of death resulting from cancers of the pancreas, large intestines, and kidney were reported. Overall risk of SNs was greater for patients who were treated with radiation therapy and chemotherapy compared with patients who were treated with radiation therapy alone, although patterns varied by organ site. In a cohort of 143 retinoblastoma survivors diagnosed between 1997 and 2006, continued improvements in mortality were seen.[4][Level of evidence C1] For patients with nonhereditary retinoblastoma, only 27 deaths in 924 patients were attributed to SNs. Among retinoblastoma survivors with heritable retinoblastoma, those with an inherited germline pathogenic variant are at a slightly higher risk of developing an SN than are those with a de novo variant. Melanoma was the most common SN seen in patients with germline pathogenic variants.[10]
Past treatment for retinoblastoma with radiation therapy. The cumulative incidence of SNs was reported to be 26% (± 10%) in nonirradiated patients and 58% (± 10%) in irradiated patients by 50 years after diagnosis of retinoblastoma, resulting in a rate of about 1% per year.[11]A German series of 633 patients with heritable retinoblastoma demonstrated a 5-year survival rate of 93%. However, 40 years later, only 80% of patients survived, with most succumbing to radiation-induced SNs (hazard ratio, approximately 3).[12] Other studies analyzing cohorts of patients treated with more advanced radiation planning and delivery technology have reported the SN rates to be about 9.4% in nonirradiated patients and about 30.4% in irradiated patients.[13] In a nonrandomized study that compared two contemporary cohorts of patients with hereditary retinoblastoma who were treated with either photon (n = 31) or proton (n = 55) therapy, the 10-year cumulative incidence of radiation-induced SNs was significantly different between the two groups (0% for proton radiation vs. 14% for photon radiation; P = .015).[14] Longer follow-up is required to further define the risk of SNs associated with proton radiation.The most common SN is sarcoma, specifically osteosarcoma, followed by soft tissue sarcoma and melanoma. These malignancies may occur inside or outside of the radiation field, although most are radiation induced. The carcinogenic effect of radiation therapy is associated with the dose delivered, particularly for subsequent sarcomas. A step-wise increase is apparent at all dose categories. In irradiated patients, two-thirds of SNs occur within irradiated tissue, and one-third of SNs occur outside the radiation field.[5,11,13,15] In a cohort of 952 irradiated survivors of hereditary retinoblastoma who were originally diagnosed between 1914 and 2006, 105 bone sarcomas and 125 soft tissue sarcomas were identified. Approximately two-thirds of these cancers occurred in the head and neck. The incidence rates were 2,000-fold higher for bone sarcomas and 500-fold higher for soft tissue sarcomas than was expected in the general population. Head and neck bone and soft tissue sarcomas were diagnosed in early childhood and continued into adulthood, with a 60-year cumulative incidence of 6.8% for bone sarcomas and 9.3% for soft tissue sarcomas. Bone and soft tissue sarcomas diagnosed elsewhere in the body were increased 169-fold and 45.7-fold, respectively, compared with the general population. Bone sarcomas primarily occurred in the long bones during adolescence. The incidence of soft tissue sarcomas was rare until age 30 years, when it rose steeply (60-year cumulative incidence, 6.6%), particularly for females (9.4%). The soft tissue sarcomas that occurred in females were leiomyosarcomas and were mainly located in the abdomen and pelvis.[5,11]; [16][Level of evidence C1]
Age at time of radiation therapy. The risk of SNs also appears to depend on the patient's age at the time that external-beam radiation therapy (EBRT) is administered, especially in children younger than 12 months. The histopathological types of SNs may be influenced by age.[13,17,18]
Previous SN. Patients who survive SNs are at a sevenfold increased risk of developing another SN.[19] The risk increases an additional threefold for patients treated with radiation therapy.[20]
The issue of balancing long-term tumor control with the consequences of chemotherapy is unresolved. Most patients who receive chemotherapy are exposed to etoposide, which has been associated with secondary leukemia in patients without a predisposition to cancer. However, most patients are exposed at modest rates when compared with the risks associated with EBRT in heritable retinoblastoma.
Despite the known increased risk of acute myeloid leukemia (AML) associated with the use of etoposide, patients with heritable retinoblastoma are not at an increased risk of developing this SN.[21-23] An initial report conducted by informal survey methods described 15 patients who developed AML after chemotherapy. One-half of the patients also received radiation therapy.[22] This finding has not been substantiated by formal studies. In a single-institution study of 245 patients who received etoposide, only 1 patient developed acute promyelocytic leukemia after 79 months.[21] Additionally, the Surveillance, Epidemiology, and End Results (SEER) Program calculated standardized incidence rates for secondary hematopoietic malignancies in 34,867 survivors of childhood cancer. The observed-to-expected ratio of secondary AML in patients treated for retinoblastoma was zero.[24]
Publication references
Read the original reference and check its publication notices.
- PubMed 11382638 · Original source
- PubMed 11581064 · Original source
- PubMed 14996857 · Original source
- PubMed 15800318 · Original source
- PubMed 17202110 · Original source
- PubMed 17613328 · Original source
- PubMed 19066271 · Original source
- PubMed 19351917 · Original source
- PubMed 19493675 · Original source
- PubMed 20400293 · Original source
- PubMed 20549819 · Original source
- PubMed 21826785 · Original source
- PubMed 22205104 · Original source
- PubMed 22355046 · Original source
- PubMed 23674091 · Original source
- PubMed 23918737 · Original source
- PubMed 24122173 · Original source
- PubMed 26255575 · Original source
- PubMed 27382102 · Original source
- PubMed 27567086 · Original source
- PubMed 30698734 · Original source
- PubMed 31622129 · Original source
- PubMed 32434110 · Original source
- PubMed 9134547 · Original source
- PubMed 9333268 · Original source
- PubMed 9544627 · Original source
Preserved source evidence · Not medical advice
Triangle