A small percentage of individuals survive their first cancer with the help of radiation yet, later in life, develop a second malignant neoplasm in the radiation port. Though it is an effective adjuvant in cancer treatment, radiation therapy is not without complications. A fraction of postradiation (1-30%) second malignant neoplasms occur in more than one site in the radiation field.13,18,19
Multifocality is defined as two or more noncontiguous, synchronous or metachronous, homopathologic or heteropathologic, postradiation malignancies occurring in the portal of a previously irradiated cancer. Theoretically, all tissue exposed to radiation is at risk for undergoing malignant transformation at some time in the future. In this text we explore the occurrence and descriptors of multi- focal post radiation sarcomas (PRS).
The hypothesis of this investigation is multifocality is a real phenomenon that needs to be addressed in the staging and management of cancer survivors who go on to develop second malignant neoplasms. In this manuscript we ask the following questions: (1) what is the frequency of multifocality in PRS; and (2) are there any predictive factors for it?
MATERIALS AND METHODS
We reviewed sarcoma databases at three institutions to identify patients who developed a postradiation second malignant neoplasm. Forty-nine individuals with postradiation malignancies were identified. One postradiation malignancy was deleted to generate a uniform population of 48 PRS (Table 1). The patients' medical records, database, imaging studies, and pathology were reviewed to collect pertinent data. Each individual was uniquely treated at only one institution. Followup ranged from 1 to 144 months with an average of 23 months. Institutional Review Board approval was obtained at each institution and all data was deidentified.
Clinic visits and phone calls were used to document the current status regarding occurrence of local recurrences or metastases. The presence or absence of two or more sarcomas within the radiation port (multifocality) was documented with biopsies and/or imaging studies, as were the other cancers that ultimately developed. Cancer 1, or preradiation malignancy, was defined as the index malignancy necessitating external beam radiotherapy. Cancer 2 is the PRS which may or may not have exhibited multifocality. Multifocality, regardless of the number of noncontiguous cancers or whether they occurred metachronously or synchronously, was still considered part of cancer 2. Cancer 3 was a third, separate malignancy that may not have been treatment related, but was outside the radiation port.
We used Kaplan Meier analysis to calculate survival with log rank comparison survival tests for the entire group, and for the multifocal versus the nonmultifocal group. Multivariate regression analysis was performed to define any predictive variables which might predispose to the development of multifocality with these variables independent, and multifocality being dependent. NCSS 2000 (Number Cruncher Statistical Systems. Kaysville, Utah 84037) was the software package used for analyses.
Multifocality occurred in 15 of the 48 (31%) patients (Table 2). Seven individuals had at least three cancers. The PRS which were multifocal were always all soft tissue sarcomas or skeletal sarcomas, but not one of each. Pathologies were similar but not necessarily identical (Figs 1-3), and we have summarized all histopathology (Table 1).
A relatively high number (10) of postmastectomy women had PRS after lumpectomy and radiation for their mammary carcinoma. Four developed postradiation skeletal sarcomas, including three osteosarcomas (two of the sternum and scapula) and one clavicular malignant fibrous histiocytoma (MFH). Six others had postradiation soft tissue sarcomas including MFH (2), leiomyosarcoma (2), liposarcoma (1) and angiosarcoma (1) at a variety of supraclavicular, chest wall, and upper extremity sites. Half(5) of the women had multifocality of their PRS in the radiation portals. Two women had third cancers. One had thyroid and lung carcinomas and leukemia (CLL), while the other had lung carcinoma.
At final followup, 19 of the 48 patients were without evidence of disease. Thirteen were alive with disease and 17 were dead of disease (Tables 3 and 4). We observed an approximate overall survival rate of 50% at 5 or 7 years, regardless of the presence or absence of multifocality (Figs 4-6, Table 5). The disease-free survival rate was approximately 30% at 5 years regardless of multifocality status. At 7.3 years, the disease-free survival rate dropped to zero for the multifocal PRS group, but this difference was not significantly different from the nonmultifocal group.
The time interval between 1° and 2° cancers (hiatus) ranged from 15 months to 672 months with a mean of 185 months (Table 5). The dose of radiation preceding the second malignant neoplasms ranged from 10 to 70Gy with a mean of 4521 cGy. The age at the time of radiation delivery ranged from 29 to 848 with a mean of 391 months (2-71, mean 33 years). We presumed these continuous variables were potential predictors of multifocal PRS tumor formation. The time hiatus variable between cancer 1 and cancer 2 predicted (p = 0.04) multifocality (Table 5). Sex, bone/soft tissue site for PRS, age and radiation dose were not shown to be predictive of multifocal PRS development.
Numerous studies5-8,10,12,15 since 1922 have reported sarcomas developing after exposure to various forms of radiation. Martland's report in 193110 described skeletal sarcomas in radium dial painters. The report of C. Howard Hatcher in 1945 of two postradiation chondrosarcomas and one fibrosarcoma established external radiation could be sarcomagenic to normal bone.7
We sought to determine how often the poorly reported entity of multifocal PRS occurred and whether or not there were any predictive variables which might help clinicians manage this particular group of cancer survivors. We note several limitations to the study. This retrospective, case- controlled series suffers from the inability to follow of large numbers of patients for a long time. Because the incidence of this occurrence is so rare, a prospective, single institution acknowledged cohort is nearly impossible. Consequently, bias is present in the study, especially in terms of all the study population identified from tertiary cancer referral centers.
Skeletal and soft tissue sarcomas, leukemia, lymphoma, gliomas, and breast and thyroid carcinomas are common second malignant neoplasms the incidence of which depends on treatment (radiation, alkylating agents, and other cytotoxic agents), gender, and patient age for the first cancer. Irradiation is associated with at least a 5 to 10-fold increase in numbers of second malignant neoplasms.8
The risk of developing a second malignant neoplasm increases if radiation therapy was given, and the incidence escalates rather than plateaus with extended followup. In one cohort of pediatric patients receiving megavolt radiation, the incidence of a second malignant neoplasm at 30 years followup is 13% with no plateau seen at extended followup.5 In a cohort of patients with acute lymphocytic leukemia (ALL) as a primary disease, there was a 0.95% incidence of a second malignant neoplasm at 20 years without radiation therapy. With radiation therapy, the incidence rose to 4% at 15 years and 20% at 20 years followup.15 The Late Effects Study Group cohort (Hodgkin's disease as the primary malignancy) reported a 10.6% incidence of second malignant neoplasms at 20 years and26.3% incidence at 30 years of followup.1 Ng et al12 provided data showing an escalating risk of developing a second malignant neoplasm at extended followup. After radiation therapy was given for Hodgkin's disease, they found a 2.3% excess risk of developing a second malignant neoplasm per year at 15 years of followup and a 4% excess risk per year at 20 years.12
Megavoltage radiation units, delivering more penetrating beams and energy higher than one megavolt (up to 60 Gy), replaced orthovoltage (250 kilovolts peak) in the mid 1960s to early 1970s.3 Megavoltage radiation has a far lower rate of malignant transformation than orthovoltage. Newer irradiation technologies, such as intensity modulated radiation (IMRT), potentially may increase the incidence of second malignant neoplasms in normal tissue when compared with conventional radiation therapy (CRT). This is because of scatter radiation as more fields are used in treatment with IMRT. This is a very small estimated increase; 1.75% incidence of a second malignant neoplasm at 10 years post treatment for IMRT compared with 1% for CRT. Also, the scatter radiation is a low dose (a few Gy); therefore, one would expect the additional second malignant neoplasms to be carcinomas as opposed to sarcomas, which are generally produced in the more heavily irradiated tissues.6 This is supported by data from the Japanese A-bomb survivors.14 Most survivors received radiation doses of less than 4 Gy. The observed rate of sarcomas in this population is no more than expected.14 In our study, 47 Gy (range, 26-70 Gy) was the mean radiation dose given to the patients who developed postradiation sarcomas. The average latency to develop postradiation sarcomas was 16 years. This data is similar to a previous series of postradiation sarcomas.2,4,11,17
Much attention has been given to the development of angiosarcoma as a second malignant neoplasm of the breast or ipsilateral upper extremity after radiation therapy in the treatment of a primary breast carcinoma. Controversy exists in the literature between prior radiation therapy and lymphedema as causative factors in the development of angiosarcoma as second malignant neoplasms. Support in the literature exists for lymphedema as having a causative role (Stewart-Treves syndrome). A Swedish-based population study found a correlation of angiosarcoma and lymphedema but found no correlation between angiosarcoma and radiation therapy.9 The SEER database supports radiation therapy as correlated with angiosarcoma as a second malignant neoplasm. The incidence of angiosarcoma as a second malignant neoplasm after breast carcinoma was higher in patients receiving radiation therapy versus no radiation therapy as a primary treatment at 15 years after diagnosis (0.9/1000 and0.1/1000, respectively).20 In our series, 10 postradiation sarcomas developed after treatment of a primary breast carcinoma; however, only one was an angiosarcoma. Our data reinforces the need to survey the entire radiation portal for any possible second malignant neoplasm. As breast conserving surgery and radiation become a more common trend for the treatment of early stage mammary carcinoma, it is possible there is a larger pool of individuals “at risk” for the development of PRS. As more patients accrue, it may affect current treatment regimens. Breast-conserving protocols for Stage I and II carcinomas are yielding similar long-term results for local and distant disease control compared with nonradiation treatments.16
Multifocality occurs in postradiation sarcomas with a frequency of 31%. There have been case reports in the literature previously alluding to the phenomenon. Tsurkan et al19 reported on the development of two simultaneous postradiation sarcomas of the abdominal wall. Tillotson et al18 reported on a case of postradiation osteosarcoma of the tibia and fibula that had four distinct foci of the neoplasm. They termed this multicentric postradiation osteosarcoma.18 Oncodera et al13 reported a double malignant neoplasm after radiation therapy for a benign head and neck pseudotumor. This was a metachronous presentation of osteogenic sarcoma and squamous cell carcinoma in the former radiation portal.13
Our data suggest multifocality may occur more frequently the longer the interval between the radiated cancer and the development of any secondary cancer. Our data did not demonstrate a survival difference between the multifocal and nonmultifocal PRS groups; however, the relatively common occurrence of approximately one in three PRS being multifocal needs to recognized by the oncology surgeon. The need to closely examine the entire radiation port as a candidate for resection should be seriously considered.
1. Bhatia S, Yasui Y, Robison LL, Bogue MK, Diller L, DeLaat C, Fossati-Bellani F, Morgan E, Oberlin O, Reaman G, Ruymann FB, Tersak J, Meadows AT, Late Effects Study Group. High risk of subsequent neoplasms continues with extended follow-up of childhood Hodgkin's disease: report from the Late Effects Study Group. J Clin Oncol
2. Bloechle C, Peiper M, Schwarz R, Schroeder S, Zornig C. Post- irradiation soft tissue sarcoma. Eur J Cancer
3. Chakravarti A, Spiro IJ, Hug EB, Mankin HJ, Efird JT, Suit HD. Megavoltage radiation therapy for axial and inoperable giant-cell tumor of bone. J Bone Joint Surg Am
4. Davidson T, Westbury G, Harmer CL. Radiation-induced soft-tissue sarcoma. Br J Surg
5. Gold DG, Neglia JP, Dusenbery KE. Second neoplasms after mega- voltage radiation for pediatric tumors. Cancer
6. Hall EJ, Wuu CS. Radiation-induced second cancers: the impact of 3D-CRT and IMRT. Int J Radiat Oncol Biol Phys
7. Hatcher CH. The development of sarcoma in bone subjected to roentgen or radium irradiation. J Bone Joint Surg Am
. 1945;27: 179-195.
8. Hawkins MM. Second primary tumors following radiotherapy for childhood cancer. Int J Radiat Oncol Biol Phys
. 1990;19: 1297-1301.
9. Karlsson P, Holmberg E, Samuelsson A, Johansson KA, WallgrenA. Soft tissue sarcoma after treatment for breast cancer-a Swedish population-based study. Eur J Cancer
10. Martland HS. The occurrence of malignancy in radioactive persons: a general review of data gathered in the study of radium dial painters, with special reference to the occurrence of osteogenic sarcoma and the interrelationship of certain blood diseases. Am J Cancer
11. Murray EM, Werner D, Greeff EA, Taylor DA. Postradiation sarcomas: 20 cases and a literature review. Int J Radiat Oncol Biol Phys
12. Ng AK, Bernardo MP, Weller E, Backstrand KH, Silver B, Marcus KC, Tarbell NJ, Friedberg J, Canellos GP, Mauch PM. Long-term survival and competing causes of death in patients with early-stage Hodgkin's disease treated at age 50 or younger. J Clin Oncol
13. Onodera K, Ichinohasama R, Ooya K. Double malignant neoplasms occurring long after local radiation to the oral mucosa. Virchows Arch
14. Pierce DA, Shimizu Y, Preston DL, Vaeth M, Mabuchi K. Studies of the mortality of atomic bomb survivors: Report 12, Part I: Cancer: 1950-1990. Radiat Res
15. Pui CH, Cheng C, Leung W. Extended follow-up of long-term survivors of childhood acute lymphoblastic leukemia. N Engl J Med
16. Santiago RJ, Harris EER, Qin L, Hwang WT, Solin LJ. Similar long term results of breast-conservation treatment for stage I and II invasive lobular carcinoma compared with invasive ductal carcinoma of the breast. Cancer
17. Sheppard DG, Libshitz HI. Post-radiation sarcomas: a review of the clinical and imaging features in 63 cases. Clin Radiol
. 2001;56: 22-29.
18. Tillotson C, Rosenberg A, Gebhardt M, Rosenthal DI. Postradiation multicentric osteosarcoma. Cancer
19. Tsurkan AM, Bogdanskii GV, Novikova NV. Simultaneous development of 2 post-radiation sarcomas of the abdominal wall. Vopr Onkol
20. Yap J, Chuba PJ, Thomas R, Aref A, Lucas D, Severson RK, Hamre M. Sarcoma as a second malignancy after treatment for breast cancer. Int J Radiat Oncol Biol Phys