|Year : 2015 | Volume
| Issue : 3 | Page : 337-342
Analyzing multiple risk factors in patients with sarcomas. A case-control study
S Nabi1, P Kahlon1, P Kuriakose2
1 Department of Internal Medicine, Henry Ford Health System, Detroit, Michigan 48202, United States
2 Department of Hematology Oncology, Henry Ford Health System, Detroit, Michigan 48202, United States
|Date of Web Publication||18-Feb-2016|
Department of Internal Medicine, Henry Ford Health System, Detroit, Michigan 48202
Source of Support: None, Conflict of Interest: None
Context: Sarcomas are a rare group of malignancies. Very little is known about their risk factors. Aims: The aim was to evaluate different risk factors in patients with sarcomas and to determine the median age at diagnosis, differences in race, gender, histological grades and staging in sarcoma patients. Settings And Design: A retrospective case-control study was conducted in a tertiary care hospital in the USA. This included patients diagnosed with sarcomas from year 2000 to 2010. Materials And Methods: Data were extracted with the help of electronic medical records using International Classification of Diseases, Ninth revision codes. Healthy, matched controls were randomly selected from the same tertiary care hospital database. Statistical Analysis: Univariate comparisons between cases and controls were done using a two-group independent t-test for age and using Chi-square tests for the categorical variables. In order to identify possible independent predictors of sarcomas, a multiple logistic regression model was constructed using sarcoma status as the dependent variable and using, initially, all variables with a univariate P < 0.2 as independent variables. Variables were reduced in a manual stepwise manner to arrive at a final model. Statistical significance was set at P < 0.05. All analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC, USA). Results: A total of 425 sarcoma patients and 429 age, sex and race matched healthy controls were analyzed in this study. We found that a history of smoking and alcoholism was significantly associated with sarcomas. We also found that the history of cancer in first-degree relatives had a significant relationship. In addition, patients with sarcomas are more likely to have a history of another malignancy when compared with controls. Conclusions: Smoking and alcohol are potential risk factors for sarcomas. In addition, a history of cancer in the first-degree relative is also a potential risk factor. Patients with sarcomas are likely to have a history of another malignancy when compared with controls.
Keywords: Alcoholism, family history, sarcoma, smoking
|How to cite this article:|
Nabi S, Kahlon P, Kuriakose P. Analyzing multiple risk factors in patients with sarcomas. A case-control study. Indian J Cancer 2015;52:337-42
| » Introduction|| |
A sarcoma (Greek origin meaning “flesh”) is a malignancy that arises from cells of mesenchymal origin. Sarcomas are quite rare, with only 15,000 new cases per year in the United States. They, therefore, represent about 1% of the 1.5 million new cancer diagnoses in the country each year. In the United States, gastrointestinal stromal tumor is the most common type of sarcoma, with 3000–3500 cases reported every year.
As sarcomas are a relatively rare group of malignancies, the literature on them is limited. The purpose of this study was to identify possible risk factors for sarcomas, which included smoking, alcoholism, family history of cancers (and their types) and personal history of another cancer (other than the primary sarcomas). Our aim was to also provide a descriptive analysis of different grades of sarcomas as well as the clinical stage at the time of diagnosis. In patients with a history of metastatic disease, our aim was to determine the common sites involved by metastases.
| » Materials and Methods|| |
This was a tertiary care hospital based case-control study in which a retrospective chart review of 854 subjects was carried out (425 patients with histologically confirmed sarcomas and 429 healthy controls matched for age, gender and race). The study was approved by our health system's institutional review board. Data were extracted from the electronic medical record system used within the health system using the International Classification of Diseases, Ninth revision coding. The variables that were analyzed included gender; race; age at the time of diagnosis; histological subtype and grade; staging based on the tumor nodes metastasis (TNM) system; history of smoking, alcoholism, and of another malignancy (in the same patients) diagnosed during adulthood; family history of cancer (in first degree relatives only) and whether or not the patient was alive at the time of the study.
Histological subtypes, as well as the grades, were analyzed based on the documentation of the senior pathologist responsible for the patient's biopsy samples. In the majority of the cases, immunohistochemistry markers, such as cytokeratins, cluster of differentiation markers and other cell surface markers, had been used by the pathologist for diagnostic confirmation. A variety of histological subtypes was identified. The grades were divided into five main types: Grade I (well differentiated), grade II (moderately differentiated), grade III (poorly differentiated), grade IV (undifferentiated) and grade X (unable to assess grade). The TNM system was used for staging. In the majority of the patients, the stage was already documented in the chart after extensive work-up involving radiological as well as invasive studies.
To standardize the smoking history, a minimum of 5 pack years were considered as a positive smoking history. Subjects with a weak or transient smoking history were considered to have a negative history in this study. The reason for this standardization was that many of our subjects had smoked one or two cigarettes a day for 1–2 years during their teenage years, which has not been proven to be significant exposure to cause carcinogenesis. Alcohol abuse was defined based on the centers for disease control and prevention's definition of more than one drink per day in females and more than two drinks per day in males. Social drinking on occasion was considered to be a negative history of alcohol abuse in our study.
Subjects' history of another malignancy other than sarcomas was also analyzed. This was based on the analysis of previous records as well as the histopathology results present in the patient's electronic medical record/chart. Both the histopathology results, as well as notes from different subjects' physicians, were analyzed carefully. Types of notes included office visits, history and physical notes and consultation notes (from subspecialties such as oncology and radiation oncology). The same notes were analyzed to determine a subject's family history of cancers. However, only first-degree relatives (parents, offspring, and siblings) were considered. Any other relatives with a history of cancer (such as aunts, uncles, grandparents) were not considered as a positive family history.
Univariate comparisons between cases and controls were done using a two-group independent t-test for age and using Chi-square tests for the categorical variables. In order to identify possible independent predictors of sarcomas, a multiple logistic regression model was constructed using sarcoma status as the dependent variable and using, initially, all variables with a univariate P < 0.2 as independent variables. Variables were reduced in a manual stepwise manner to arrive at a final model. Statistical significance was set at P < 0.05. All analyses were performed using SAS 9.4 (SAS Institute Inc. Cary, NC, USA).
| » Results|| |
A total of 854 subjects were analyzed. In the univariate analyses, a history of smoking (33% vs. 13%), alcoholism (15% vs. 4%), a history of a second malignancy (21% vs. 9%) and family history of cancer (35% vs. 12%) was significantly higher in sarcoma patients as compared to controls. Among the first degree relatives, the ones with statistical significance during univariate analyses were mothers (19% vs. 5%), fathers (13% vs. 4%), brothers (4% vs. 2%) and sisters (8% vs. 2%). In a multivariate logistic regression analyses, the factors that remained statistically significant were smoking, alcoholism and a family history of cancer, that is, in fathers, mothers and sisters [Table 1].
|Table 1: Descriptive statistics with univariate and multivariate analysis|
Click here to view
The median age at the time of diagnosis was 54.7 years. Minimum age at diagnosis was 4 years, and a maximum was 96 years. Grade III was the most common histological grade, identified in 207 (49%) patients, followed by grade I in 96 (23%) patients, grade II in 62 (15%) patients and grade IV in 16 patients (4%). Forty-four (10%) patients had grade X. At least 24 histological subtypes of sarcomas were present in our patient population [Table 2]. Among the histological subtypes, leiomyosarcoma was found to be the most common, and was found in 63 (14%) patients, followed by spindle cell sarcoma in 56 (13%) patients.
Stage four was found to be the most common stage, comprising 187 (44%) of the patients, followed by stage one in 104 (24%) patients, stage two in 86 (20%) patients, and the least common was stage three in 48 (11%) patients. The most common location of primary tumors was the pelvis in 22% patients, followed by the lower extremities in 19% and the abdomen in 17% [Table 3]. Total number of sites involved in metastases was 228 (in 187 patients). The most common location involved by metastatic disease was the chest (39%), followed by the abdomen (29%) and the pelvis (9%). In the chest, lungs were the most common site of metastases [Table 4].
Eighty-eight (21%) of our sarcoma patients and 38 (9%) controls had a history of another malignancy diagnosed during adulthood. The total number of malignancies (other than sarcoma) in this subgroup of 88 patients was 100. The most common malignancies among them were breast cancer (19%), basal cell cancer (15%), and prostate cancer (15%) [Table 5].
|Table 5: Location of second primary malignancy in patients with sarcoma (n=100)|
Click here to view
Interestingly, 147 (35%) of our sarcoma patients and 53 (12%) of the controls had one or more first degree relatives with a history of cancer. Of the 147 sarcoma patients, 51 (35%) patients had two or more first degree relatives with cancer and the remaining 96 (65%) patients each had one first degree relative with cancer. There were a total of 193 first degree relatives (of our patient population) with a history of cancer. The most common first degree relative to have a cancer was the mother (38%), followed by the father (30%) [Figure 1]. There were a total of 208 cancers in the 193 first degree relatives (a few relatives had more than one cancer). The most common cancer was breast cancer (19%), followed by lung cancer (16%) and colorectal cancer (12%) [Table 6].
|Table 6: Types of cancer in first degree relatives of patients with sarcoma (n=208)|
Click here to view
At the time of our study, 167 (40%) patients from our total population were deceased.
| » Discussion|| |
The aim of this study was to analyze possible risk factors in a large population of sarcoma patients. The variables included age at the time of diagnosis, gender and race distribution, major histological subtypes, clinical stage of the disease at the time of presentation and common metastatic sites in patients with stage four disease. Other important variables included a strong smoking history, history of alcohol abuse, history of another primary cancer diagnosed during adulthood, first degree relatives of our patients with cancer, types of cancer in the first degree relatives and whether or not the patient was alive at the time of our study.
Two important variables analyzed in this study were smoking and alcoholism. Smoking has a proven association with a number of malignancies including lung cancer, colon cancer, and bladder cancer. A large number of carcinogens in cigarette smoke have been implicated as contributors to oncogenesis. Review of the literature has shown a positive association between smoking and sarcomas in a few studies. However, the amount of evidence is limited. A case-control study by Zahm et al. showed a positive association between sarcomas, smoking and chewed tobacco. A prospective study by Zahm et al. showed a positive association between smoking and sarcomas in United States veterans. Interestingly, the risk was even higher among former smokers. Our study showed a statistically significant association between smoking and sarcomas, and hence, supports such an association. A similar statistically significant association was found between alcohol and sarcomas in our study. To the authors' knowledge, this is the first study to report such an association. There is also evidence to suggest that certain occupational factors may have a relationship with sarcomas. Occupations involving the usage of pesticides, herbicides and insecticides were found to have an association with sarcomas.
More than 20% of our patients had a history of another malignancy diagnosed during adulthood. This raises the question about inheritance and its role in sarcoma development. A few inherited conditions are associated with sarcomas. Patients with Li–Fraumeni syndrome, an autosomal dominant condition, have an approximately 25-fold increased risk of developing a malignant tumor by age 50 than the population average and are at risk for a wide range of malignancies, with particularly high occurrences of soft tissue sarcomas, breast cancer, brain tumors, acute leukemia, bone sarcomas, and adrenal cortical carcinoma.,, The review of the literature shows that many case reports of Li-Fraumeni syndrome have been published., This syndrome results from a mutation in the p53 gene, which is a tumor suppressor gene involved in DNA repair. Another similar association is found between retinoblastomas and osteosarcomas due to mutation in the RB-1 tumor suppressor gene. In addition to the risk for osteosarcoma, a more recent study highlights that survivors of hereditary retinoblastoma are also at an increased risk of developing soft-tissue sarcomas within the radiation field, especially leiomyosarcoma. Koivisto-Korander et al. concluded that the risk of primary cancers (such as breast and colon cancer) increased in females with uterine leiomyosarcoma. Our results showed that patients with sarcomas are more likely to have other cancers as compared to controls. Possible mechanisms include genetic/acquired mutations leading of carcinogenesis.
In terms of family history, we analyzed only the first degree relatives with one or more cancers. In our patient population, 35% of patients had a family history of cancer which was statistically significant when compared with controls. Of these patients, 35% had two or more first degree relatives with cancer, and the remaining 65% had a single relative with cancer. McDuffie et al. suggests that patients with soft tissue sarcomas, lymphoma and multiple myeloma are more likely to have a first degree relative with cancer as compared to controls. In our study, the most common first degree relative to have a cancer history was the mother. A recent article suggests that a reduced number of mitochondrial DNA copies in peripheral blood leukocytes is associated with the development of soft tissue sarcomas. As mitochondrial DNA is maternally inherited, this raises the question whether soft tissue sarcomas can be maternally inherited. However, to the authors' knowledge, this association has never been reported in the literature. Further studies are needed to determine the pattern of inheritance of different types of sarcomas and to understand the genetics better.
The median age at the time of diagnosis was approximately 54 years, and 70% of our patient population was between the ages of 41 and 80 years. This pattern is consistent with the pattern of most solid tumors, such as colon cancer, lung cancer, and breast cancer. It is well known that the risk of carcinogenesis increases with the ageing process, especially in patients with continuous carcinogen exposure, such as cigarette smoking in lung cancers. Also, there might be a role of declining DNA repair mechanisms with advancing age, which can also contribute to carcinogenesis. In terms of gender difference, our population was found to have no predominance towards males or females. To the authors' knowledge, no specific gender predominance has been reported in sarcoma patients in the past. In our study, Caucasians comprised the biggest proportion of our patient population (56%) which was significantly higher than African Americans (38%). To our knowledge, in terms of sarcoma prevalence, no significant racial differences have been reported in the past. However, in terms of prognosis, African American race has been shown to be associated with a worse prognosis in patients with uterine carcinosarcoma and leiomyosarcomas.,,
In terms of histological grades of the soft tissue sarcomas, the general consensus is that higher grades are typically associated with a poor prognosis. Studies have shown that the presence of a higher grade is also associated with an increased risk of late metastatic recurrence. In our study, the most common grade was found to be grade III, that is poorly differentiated neoplasm, which roughly correlates with the most common clinical stage in our study, that is stage four. The association between grading and prognosis is also supported by the fact that the TNM staging of sarcomas also utilizes tumor grade in determining the clinical stage.
Low-grade sarcomas are usually treated with surgery alone (if no systemic metastases are found). High-grade sarcomas are generally associated with a higher risk of metastasis and worse prognosis. Therefore, they are usually treated with a combination of surgery and chemotherapy with or without radiation therapy. The recognition that many sarcomas are sensitive to chemotherapy has dramatically improved the survival of patients. For example, in the era before chemotherapy, long term survival for patients with localized osteosarcoma was only approximately 20%, but now has risen to 60–70%.
The main limitation of this study is that it is a single center study done in a tertiary care facility and the results might not be applicable to the general population. A possible bias in this study is that patients with cancer are more likely to remember a cancer history in their family members as compared to controls. However, every effort was made to gather the data as clearly as possible and to avoid bias.
| » Conclusion|| |
Our study confirms an association between smoking and alcoholism with sarcomas. Patients with a first degree relative with cancer are more likely to develop sarcomas as compared to controls. Furthermore, a previous history of cancer is also a risk factor for the development of sarcomas later in life.
| » References|| |
Borden EC, Baker LH, Bell RS, Bramwell V, Demetri GD, Eisenberg BL, et al.
Soft tissue sarcomas of adults: State of the translational science. Clin Cancer Res 2003;9:1941-56.
Tran T, Davila JA, El-Serag HB. The epidemiology of malignant gastrointestinal stromal tumors: An analysis of 1,458 cases from 1992 to 2000. Am J Gastroenterol 2005;100:162-8.
Hoffmann D, Hoffmann I. The changing cigarette, 1950-1995. J Toxicol Environ Health 1997;50:307-64.
Zahm SH, Blair A, Holmes FF, Boysen CD, Robel RJ, Fraumeni JF Jr. A case-control study of soft-tissue sarcoma. Am J Epidemiol 1989;130:665-74.
Zahm SH, Heineman EF, Vaught JB. Soft tissue sarcoma and tobacco use: Data from a prospective cohort study of United States veterans. Cancer Causes Control 1992;3:371-6.
Merletti F, Richiardi L, Bertoni F, Ahrens W, Buemi A, Costa-Santos C, et al.
Occupational factors and risk of adult bone sarcomas: A multicentric case-control study in Europe. Int J Cancer 2006;118:721-7.
Mitchell G, Ballinger ML, Wong S, Hewitt C, James P, Young MA, et al.
High frequency of germline TP53 mutations in a prospective adult-onset sarcoma cohort. PLoS One 2013;8:e69026.
Curry S, Ibrahim F, Grehan D, McDermott M, Capra M, Betts D, et al.
Rhabdomyosarcoma-associated renal cell carcinoma: A link with constitutional Tp53 mutation. Pediatr Dev Pathol 2011;14:248-51.
Schneider K, Zelley K, Nichols KE, Garber J. Li-Fraumeni syndrome. In: Pagon RA, Adam MP, Ardinger HH, Bird TD, Dolan CR, Fong CT, et al
., editors. GeneReviews ®
. Seattle WA: University of Washington, Seattle; 1993.
Sui X, Li Y, Zhao H, Wang J. Giant liposarcoma of the esophagus with Li-Fraumeni-like syndrome. Eur J Cardiothorac Surg 2011;40:1253-5.
Henry E, Villalobos V, Million L, Jensen KC, West R, Ganjoo K, et al.
Chest wall leiomyosarcoma after breast-conservative therapy for early-stage breast cancer in a young woman with Li-Fraumeni syndrome. J Natl Compr Canc Netw 2012;10:939-42.
Friend SH, Bernards R, Rogelj S, Weinberg RA, Rapaport JM, Albert DM, et al.
A human DNA segment with properties of the gene that predisposes to retinoblastoma and osteosarcoma. Nature 1986;323:643-6.
Kleinerman RA, Tucker MA, Abramson DH, Seddon JM, Tarone RE, Fraumeni JF Jr. Risk of soft tissue sarcomas by individual subtype in survivors of hereditary retinoblastoma. J Natl Cancer Inst 2007;99:24-31.
Koivisto-Korander R, Scélo G, Ferro G, Mellemkjaer L, Hemminki K, Weiderpass E, et al.
Second primary malignancies among women with uterine sarcoma. Gynecol Oncol 2012;126:30-5.
McDuffie HH, Pahwa P, Karunanayake CP, Spinelli JJ, Dosman JA. Clustering of cancer among families of cases with Hodgkin Lymphoma (HL), Multiple Myeloma (MM), Non-Hodgkin's Lymphoma (NHL), Soft Tissue Sarcoma (STS) and control subjects. BMC Cancer 2009;9:70.
Xie H, Lev D, Gong Y, Wang S, Pollock RE, Wu X, et al.
Reduced mitochondrial DNA copy number in peripheral blood leukocytes increases the risk of soft tissue sarcoma. Carcinogenesis 2013;34:1039-43.
Benson D, Mitchell N, Dix D. On the role of aging in carcinogenesis. Mutat Res 1996;356:209-16.
Erickson BK, Doo DW, Zhang B, Huh WK, Leath CA 3rd
. Black race independently predicts worse survival in uterine carcinosarcoma. Gynecol Oncol 2014;133:238-41.
Garg G, Kruger M, Christensen C, Deppe G, Toy EP. Stage III uterine carcinosarcoma: 2009 international federation of gynecology and obstetrics staging system and prognostic determinants. Int J Gynecol Cancer 2011;21:1606-12.
Kapp DS, Shin JY, Chan JK. Prognostic factors and survival in 1396 patients with uterine leiomyosarcomas: Emphasis on impact of lymphadenectomy and oophorectomy. Cancer 2008;112:820-30.
Kang S, Kim HS, Kim S, Kim W, Han I. Post-metastasis survival in extremity soft tissue sarcoma: A recursive partitioning analysis of prognostic factors. Eur J Cancer 2014;50:1649-56.
Toulmonde M, Le Cesne A, Mendiboure J, Blay JY, Piperno-Neumann S, Chevreau C, et al.
Long-term recurrence of soft tissue sarcomas: Prognostic factors and implications for prolonged follow-up. Cancer 2014;120:3003-6.
Longhi A, Errani C, De Paolis M, Mercuri M, Bacci G. Primary bone osteosarcoma in the pediatric age: State of the art. Cancer Treat Rev 2006;32:423-36.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]