Papers

Cancers coinciding with childbearing: delayed diagnosis during pregnancy?

^a Department of Cancer Epidemiology, University Hospital, S-751 85 Uppsala, Sweden

Correspondence to: Dr Lambe. e-mail: mats.crabix{at}epic.uu.se.

Cancers associated with childbearing pose extremely difficult questions for the patient and her physician. Population based registry data gave us the opportunity to examine the incidence of this uncommon occurrence.

Subjects, methods, and results

Cases were identified following a linkage between the Swedish Cancer Registry and a nationwide fertility registry. Nearly all diagnosed cancers are recorded in the cancer registry.¹ The fertility registry contains information on number and dates of live births for more than 2.3 million Swedish women born in 1925 and thereafter.²

Among women born from 1925 to 1972 more than 2.7 million live births and 32848 cancers were recorded during reproductive ages (15-44 years) in 1960-90. Of all cancers, 428 (1.3%) were diagnosed during pregnancy (date of a live birth-9 months) and 1425 (4.3%) during the lactation period (date of birth+12 months). The overall incidences during pregnancy and lactation, respectively, were 15.6 and 51.6 per 100000 live births.

Median age at diagnosis during pregnancy was 29.8 years. The most frequent sites were skin (malignant melanoma; 3.6 per 100000 live births), cervix uteri (2.4 per 100000), and breast (2.0 per 100000). About 4% of breast cancers and 10% of thyroid cancers during reproductive years were diagnosed during pregnancy or in the year after birth. The corresponding figures in the age group 25-29, when childbearing was most frequent, were 19% and 21%, respectively.

Expected numbers during pregnancy and lactation were estimated from female age specific and period specific population rates. The observed to expected ratios during pregnancy were below unity for all of the 10 most common sites except melanoma. For all sites combined the observed to expected ratio was 0.4(9) [0.4(4)-0.5(3)]. For sites where case interval numbers indicated a deficit of recorded cases in early pregnancy (cervix uteri, breast, overy, Hodgkin's disease, leukaemias), ratios were also computed based on observed numbers in trimester 3. During the lactation period the observed to expected ratio for all sites combined was 1.2(1) [1.1(5)-1.2(8)].

Comment

Our findings are in broad agreement with a previous population based study that observed fewer than expected cancers during pregnancy.³ Our data indicate that a small, but not negligible, proportion of all malignancies in young women are diagnosed in association with childbearing. More importantly, the estimated observed to expected ratios suggest that diagnosis is delayed to the postpartum period. Pregnancy is usually a period of intense medical observation. However, potentially harmful diagnostic procedures are probably less likely to be implemented. Moreover, unusual signs and symptoms may be interpreted as being related to pregnancy. Given the physiological changes in the breast during childbearing this may be most evident for breast cancer; longer delays in pregnant compared with non-pregnant subjects have been reported.⁴ In our data, fewer cases of breast cancer were diagnosed in the first half than in the second half of the lactation period.

An alternative explanation, but not mutually exclusive, is that tumour progression is altered during pregnancy and lactation. The growth of malignant cells, present before conception, may be stimulated by transient hormonal⁵ and perhaps immunological changes during childbearing.

A limitation of our data was the lack of information on terminated pregnancies, which may have reduced the number of registered cases in the first and second trimesters. The practice of therapeutic abortions in association with malignant disease has varied according to site, stage, type of treatments available, time, and period of diagnosis. The only site where organised screening activities may have affected results is the cervix. Smear testing practices in early pregnancy have varied over time.

Ten most common sites of cancers coinciding with childbearing: observed (O) and expected (E) numbers and ratios (and
95% confidence intervals) of these sites during pregnancy (trimesters 1-3) and lactation. Expected numbers based on
general female population rates
-----------------------------------------------------------------------------------------------------------------------------------------
                                             During pregnancy*                                  Lactation period*
-----------------------------------------------------------------------------------------------------------------------------------------
                                               No of cases                                     No of cases
---------------------------------------------------------------------    Observed:  ------------------------------   Observed:
                                                                         expected           Observed                 expected
ICD-7                                       Observed                  (95% confidence     (first half+            (95% confidence
No            Site                    1++  2++  3++  Total Expected      interval)        second half)  Expected     interval)
-----------------------------------------------------------------------------------------------------------------------------------------
190       Melanoma                    26   35   39    100    90.3    1.11 (0.88 to 1.32)   171 (92+79)    120.5   1.42 (1.21 to 1.63)
171       Cervix uteri                15   16   34     65   158.0    0.41 (0.31 to 0.51)   213 (121+92)   210.7   1.01 (0.88 to 1.15)
          Cervix uteri-adjusted+                      102   158.0    0.65 (0.52 to 0.77)
170       Breast                       4   13   38     55   156.5    0.35 (0.26 to 0.44)   248 (89+159)   208.6   1.19 (1.04 to 1.34)
          Breast-adjusted+                            114   156.5    0.73 (0.59 to 0.85)
193       Nervous system              12   10   10     32    71.7    0.45 (0.29 to 0.60)   129 (66+63)     95.6   1.35 (1.12 to 1.58)
175       Ovary                        1    8   18     27    62.6    0.43 (0.27 to 0.59)    70 (35+35)     83.4   0.84 (0.64 to 0.96)
          Ovary-adjusted+                              54    62.6    0.86 (0.63 to 1.09)
153       Colon                        6   10    8     24    31.9    0.75 (0.43 to 1.05)    57 (27+30)     42.3   1.35 (1.00 to 1.70)
195       Endocrine glands             7    9    4     20    33.2    0.60 (0.34 to 0.87)    46 (20+26)     44.2   1.04 (0.68 to 1.24)
201       Hodgkin's disease            3    6    8     17    31.2    0.54 (0.29 to 0.80)    58 (29+29)     41.7   1.40 (1.03 to 1.75)
          Hodgkin's-adjusted+                          24    31.2    0.77 (0.46 to 1.07)
194       Thyroid                     10    3    4     17    50.0    0.34 (0.18 to 0.50)   118 (54+64)     66.7   1.77 (1.44 to 2.08)
204-208   Leukaemia                    0    1    8      9    29.1    0.31 (0.11 to 0.49)    39 (16+23)     38.7   1.01 (0.23 to 1.22)
          Leukaemia-adjusted+                          24    29.1    0.82 (0.49 to 1.15)
140-209   All sites                   94   130  204    428   880.7   0.49 (0.44 to 0.53)  1425 (714+711)  1174.2  1.21 (1.15 to 1.28)
-----------------------------------------------------------------------------------------------------------------------------------------
*Pregnancy period: date of delivery-9 months; lactation period: following 12 months.
+Because of possible underreporting of cases in trimesters 1 and 2 owing to induced abortions following diagnosis, O:E ratios were also
calculated based on observed numbers in trimester 3--for example, breast cancer 38 (number of cases observed in trimester 3)x3=114.
++Trimester.

Our findings may have implications for maternity care policies. Gestational cancer is likely to become more common as women in many Western countries postpone childbearing to an older age. Better awareness among doctors, coupled with an increased readiness to evaluate unusual symptoms, may help avoid undue delay in diagnosis in this special group of patients.

We are indebted to Mr Hakan Jansson for programming assistance.

Funding: Supported by grants from the Swedish Cancer Society and Wahlmarks fund at the Uppsala City Council.

Conflict of interest: None.

1. Swedish Cancer Registry. Cancer incidence in Sweden, 1990. Stockholm: National Board of Health and Welfare, Cancer Registry, 1993.
2. Johansson L, Finnas F. Fertility of Swedish women born 1927-1960. No 14 of Urval. Stockholm: Statistics Sweden, 1983.
3. Haas JF. Pregnancy in association with a newly diagnosed cancer: a population based epidemiological assessment. Int J Cancer 1984;34:229-35. [Medline]
4. Wallack MK, Wolf JA, Bedwinck J, Denes AE, Glasgow G, Kumar B, et al. Gestational carcinoma of the female breast. Curr Probl Cancer 1983;7:1-58.
5. Musey VC, Collins DC, Brogan DR, Santo VR, Musey PI, Martino-Saltzman D, Preedy JRK. Long term effects of a first pregnancy on the hormonal environment: estrogens and androgens. J Clin Endocr Metab 1986;64:111-7. [Abstract/Free Full Text]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    StumbleUpon    Technorati    What's this?

This article has been cited by other articles:

• Stensheim, H., Moller, B., van Dijk, T., Fossa, S. D. (2009). Cause-Specific Survival for Women Diagnosed With Cancer During Pregnancy or Lactation: A Registry-Based Cohort Study. JCO 27: 45-51 [Abstract] [Full text]  
• Wohlfahrt, J., Andersen, P. K., Mouridsen, H. T., Melbye, M. (2001). Risk of Late-stage Breast Cancer after a Childbirth. Am J Epidemiol 153: 1079-1084 [Abstract] [Full text]