Cancer risk in women with previous breast cancer

F. Levi1,2,+, V.-C. Te1, L. Randimbison1 and C. La Vecchia1,3,4

1 Cancer Epidemiology Unit and Cancer Registry of Vaud, Institut Universitaire de Médecine Sociale et Préventive, Lausanne; 2 Cancer Registry of Neuchâtel, Neuchâtel, Switzerland; 3 Laboratory of Epidemiology, Istituto di Ricerche Farmacologiche ‘Mario Negri’, Milan; 4 Istituto di Statistica Medica e Biometria, Università degli Studi di Milano, Milan, Italy

Received 5 August 2002; accepted 28 August 2002


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background:

Excess risks of several second neoplasms following breast cancer have been reported. However, these risks have still to be quantified.

Patients and methods:

We considered 9729 breast cancer patients registered by the Swiss Cancer Registries of Vaud and Neuchâtel (covering about 786 000 inhabitants) and followed up from 1974 to 1998.

Results:

Overall, 443 second primary neoplasms (other than second primary breast cancers) were observed versus 389 expected [standardised incidence ratio (SIR): 1.14; 95% confidence interval (CI) 1.04–1.25]. The SIRs were above unity for endometrium (SIR = 1.5), ovary (1.3), colorectum (1.1), gallbladder (1.4), cutaneous malignant melanoma (1.4), kidney (1.4), lymphomas (1.4) and leukaemias (1.2), as well as for selected tobacco-related neoplasms. The largest excess risk was found for soft tissue sarcomas (STS) with 10 cases observed versus 3.1 expected (SIR = 3.2; 95% CI 1.5–5.9). Of these, eight occurred in potentially irradiated areas.

Conclusions:

This analysis confirms the existence of a modest excess in several neoplasms occurring after breast cancer. The substantial excess of STS confirms the strong association between irradiation and STS.

Key words: breast cancer, follow-up study, population-based, second primary cancer, soft-tissue sarcoma, standardised incidence ratio


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Excess risks of second neoplasms following breast cancer, including second primary breast cancer, but also lung, melanoma and non-melanoma skin cancers, endometrial, ovarian and thyroid cancers, were found in studies in Israel, Japan and Slovenia [13]. Other studies in Sweden and Germany found no appreciable excess of neoplasms, other than breast, following breast cancer [4, 5].

A large study of 141 053 women with invasive breast cancer registered in Sweden from 1958 to 1997, and followed-up to December 1997, found an excess risk of lung cancer [standardised incidence ratio (SIR) = 1.32] [6] following breast cancer. The risk tended to rise with increasing time since breast cancer diagnosis and was attributed to an interaction between radiotherapy and smoking habits in subsequent cohorts of women.

We considered second primary cancers in breast cancer patients diagnosed in Vaud from 1974 to 1994 [7]. To provide further information on the issue of second primary neoplasms of the lung and other sites, we can now produce a similar analysis of the Neuchâtel Cancer Registry together with the Vaud Cancer Registry, and extend the follow-up to 1998.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The datasets of the Vaud and Neuchâtel Cancer Registries were considered, which include data concerning incident cases of malignant neoplasms in the Cantons, whose populations, according to the 31 December 2000 census, were 620 294 and 165 731 inhabitants, respectively. Population-based incidence data have been available since 1974. The registries are tumour based and multiple primaries in the same person are registered separately. Both passive and active follow-up are recorded, and each subsequent item of information is used to complete the records of the patient [8, 9].

After exclusion of 29 cases detected at autopsy, 21 at death, 161 by death certification alone, 845 followed-up for <1 month and of synchronous cancers (i.e. within 1 month after the first primary; n = 37), the present series comprised 9729 invasive breast cancers diagnosed from 1974 to 1998 (rate of histological verification: 97.8%). These women were followed-up to the end of 1998 for the occurrence of second primary neoplasms, emigration or death, for a total of 61 834 person-years at risk. Calculation of expected number of cases was based on register, site, age and calendar period specific incidence rates, multiplied by the corresponding number of person-years at risk. The significance of the observed:expected ratios (SIR) and the corresponding 95% confidence intervals (CI) was based on the Poisson distribution.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Table 1 gives the observed and expected number of several cancers (other than second primary breast cancers), and the corresponding SIRs, overall and in separate strata of time since breast cancer diagnosis. There were 34 lung cancers following breast cancer, corresponding to an SIR of 1.04. For lung cancer, as well as other major tobacco-related neoplasms, including oesophagus and urinary bladder cancer, the SIR tended to rise with time since breast cancer diagnosis. Among the other neoplasms showing SIRs above unity were endometrium (SIR = 1.47), ovary (1.26), colorectum (1.14), gallbladder (1.44), cutaneous malignant melanoma (1.41), kidney (1.38), lymphomas (1.38) and leukaemias (1.20).


View this table:
[in this window]
[in a new window]
 
Table 1. Observed (O) and expected (E) cases of subsequent cancer sites after an initial diagnosis of breast invasive cancer, and corresponding overall standardised incidence ratios (SIRs) and 95% confidence intervals (CI) (Vaud and Neuchâtel, Switzerland, 1974–1998)
 
The largest excess risk, however, was found for soft tissue sarcomas (STS), with 10 cases observed versus 3.1 expected (SIR = 3.20; 95% CI 1.53–5.89). The SIR was 6.1 (95% CI 2.8–11.5) >=5 years after breast cancer diagnosis. Of the 10 cases of STS registered, six occurred in the thorax or shoulder and two in the pelvis. Age ranged from 47 to 89 years (overall median, 64 years; 63 years among the eight cases in potentially irradiated areas). The histologies were as follows: four hemangiosarcomas, two fibrosarcomas, one pleiomorphic sarcoma and one leiomyosarcoma occurring in potentially irradiated areas and two leiomyosarcomas in other areas (i.e. lower limbs).

Overall, 443 second primary neoplasms were observed versus 389 expected, corresponding to an SIR of 1.14 (95% CI 1.04–1.25). The SIR was 1.04 for <5 years and 1.25 (95% CI 1.09–1.42) for >=5 years since the original diagnosis of breast cancer.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The present update analysis confirms the existence of an excess in several neoplasms following a diagnosis of breast cancer. There was also a modest excess in lung and other tobacco-related neoplasms occurring a few years after breast cancer. The excess is, however, appreciably smaller than that registered in Sweden [6], reflecting the later calendar period of the present series—and hence the use of more modern radiotherapy—and the different smoking patterns of Swiss as compared with Scandinavian women [10].

The main finding of the present analysis is, however, a substantial excess of STS in women diagnosed with breast cancer. An excess of breast cancer following STS has been reported in North West England [11]. Common correlates of breast cancer and STS, including reproductive factors [12] and overweight [13], may partly explain the excess risk. Most STSs, however, occurred in the thorax, shoulder and pelvis, i.e. in irradiated areas, and are probably the late consequences of radiotherapy, which could be documented through the cancer registries’ files for five out of eight cases located in potentially irradiated areas, and for none out of the two remaining cases. The age distribution of breast and second primary cancers is inconsistent with the classic Li–Fraumeni syndrome [14]. They confirm, nonetheless, the strong association between irradiation and STS, which appears to be mediated by variable susceptibility (i.e. germ-line mutation of the tumour suppressor gene p53) in family members [14].


    Acknowledgements
 
This work was supported by the Swiss League against Cancer.


    Footnotes
 
+ Correspondence to: Dr Fabio Levi, Cancer Epidemiology Unit and Cancer Registry of Vaud, Institut Universitaire de Médecine Sociale et Préventive, CHUV-Falaises 1, 1011 Lausanne, Switzerland. Tel: +41-21-314-73-11; Fax: +41-21-323-03-03; E-mail: fabio.levi{at}inst.hospvd.ch Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Schenker JG, Levinsky R, Ohel G. Multiple primary malignant neoplasms in breast cancer patients in Israel. Cancer 1984; 54: 145–150.[ISI][Medline]

2. Murakami R, Hiyama T, Hanai A, Fujimoto I. Second primary cancers following female breast cancer in Osaka, Japan: a population-based cohort study. Jpn J Clin Oncol 1987; 17: 293–302.[ISI][Medline]

3. Volk NV, Pompe-Kirn V. Second primary cancers in breast cancer patients in Slovenia. Cancer Causes Control 1997; 8: 764–770.[CrossRef][ISI][Medline]

4. Adami HO, Bergkvist L, Krusemo U, Persson I. Breast cancer as a risk factor for other primary malignant diseases. A nationwide cohort study. J Natl Cancer Inst 1984; 73: 1049–1055.[ISI][Medline]

5. Brenner H, Siegle S, Stegmaier C, Ziegler H. Second primary neoplasms following breast cancer in Saarland, Germany, 1968–1987. Eur J Cancer 1993; 29: 1410–1414.[CrossRef]

6. Prochazka M, Granath F, Ekbom A et al. Lung cancer risks in women with previous breast cancer. Eur J Cancer 2002; 368: 1520–1525.[CrossRef]

7. Levi F, Randimbison L, Te VC, La Vecchia C. Second primary cancers in breast cancer patients in Vaud, Switzerland. Cancer Causes Control 1998; 9: 463–464.[CrossRef][ISI][Medline]

8. Levi F, Randimbison L, Te VC et al. Multiple primary cancers in the Vaud Cancer Registry, Switzerland, 1974–1989. Br J Cancer 1993; 67: 391–395.[ISI][Medline]

9. Levi F, Portal I, Méan AM et al. Statistics from the Registry of the Canton of Neuchâtel, Switzerland, 1988–1992. In Parkin DM, Whelan S, Ferlay J et al. (eds): Cancer Incidence in Five Continents, Vol. VII. IARC Scientific Publications No. 120. Lyon: IARC 1997; 674–677.

10. Franceschi S, Naett C. Trends in smoking in Europe. Eur J Cancer Prev 1995; 4: 271–284.[ISI][Medline]

11. Hartley AL, Blair V, Harris M et al. Multiple primary tumours in a population-based series of patients with histopathologically peer-reviewed sarcomas. Br J Cancer 1993; 68: 1243–1246.[ISI][Medline]

12. Fioretti F, Tavani A, Gallus S et al. Menstrual and reproductive factors and risk of soft tissue sarcomas. Cancer 2000; 88: 786–789.[CrossRef][ISI][Medline]

13. Tavani A, Soler M, La Vecchia C et al. Body weight and risk of soft-tissue sarcoma. Br J Cancer 1999; 81: 890–892.[CrossRef][ISI][Medline]

14. Zahm SH, Tucker MA, Fraumeni JF Jr. Soft tissue sarcomas. In Schottenfeld D, Fraumeni J Jr (eds). Cancer Epidemiology and Prevention. Oxford: Oxford University Press 1996; 984–999.