Relation of risk of contralateral breast cancer to the interval since the first primary tumour.

BACKGROUND: There is no consensus on how to separate contralateral breast cancer (CBC) occurring as distant spread of the primary breast cancer (BC) from an independent CBC. METHODS: We used standardised incidence ratios (SIRs) to analyse the variations in the risk of CBC over time among 6629 women with BC diagnosed between 1954 and 1983. To explore the most appropriate cutoff to separate the two types of CBC, we analysed the deviance between models including different cutoff points as compared with the basal model with no cutoff date. We also performed a prognostic study through a Cox model. RESULTS: The SIR was much higher during the first 2 years of follow-up than afterwards. The best cutoff appeared to be 2 years. The risk of early CBC was linked to tumour spread and the risk of late CBC was linked to age and to the size of the tumour. Radiotherapy was not selected by the model either for early or late CBC risk. CONCLUSION: A clearer pattern of CBC risk might appear if studies used a similar cutoff time after the initial BC.

Consumption of dairy produce and alcohol in a case-control study of breast cancer.

In a French case-control study of 1,010 breast cancer cases and 1,950 controls with nonmalignant diseases, the risk of breast cancer was found to be positively associated with frequency of cheese consumption and the level of fat in the milk consumed. A negative association was found between frequency of yogurt consumption and the risk of breast cancer. No association was found between the consumption of butter and the risk of breast cancer. The positive association between a daily consumption of alcohol and the risk of breast cancer previously reported was not altered when dairy produce consumption was taken into account.

Conservative treatment versus mastectomy in early breast cancer: patterns of failure with 15 years of follow-up data. Institut Gustave-Roussy Breast Cancer Group.

PURPOSES: A randomized trial was conducted to compare tumorectomy and breast irradiation with modified radical mastectomy. We have analyzed the patterns of failure in each arm of the trial and the prognostic factors that have an independent effect on treatment failures and overall survival. PATIENTS AND METHODS: The trial included 179 patients with breast cancer of up to 20 mm in diameter at macroscopic examination. Eighty-eight patients had conservative management and 91 a mastectomy. All patients had axillary dissection with frozen-section examination. For patients with positive axillary nodes (N+), a second randomization was performed: lymph node irradiation versus no further regional treatment. Patterns of failure were determined by a competing-risk approach and multivariate analysis. A prognostic-score was determined by multivariate analysis. RESULTS: Overall survival, distant metastasis, contralateral breast cancer, new primary malignancy, and locoregional recurrence rates were not significantly different between the two surgical groups, or between lymph node irradiation groups. Most recurrences appeared during the first 10 years. Three distinct prognostic groups were determined taking into account age, tumor size, histologic grading, and number of positive axillary nodes. CONCLUSION: Long-term results support conservative treatment with limited surgery and systematic breast irradiation as a safe procedure for the management of small breast cancers. Four easily obtainable clinical and histologic factors may be combined in a prognostic score that is highly predictive of overall and event-free survival.

Prognostic value of c-myc proto-oncogene overexpression in early invasive carcinoma of the cervix.

The prognostic effect of c-myc oncogene overexpression was assessed in a multivariate analysis of 93 patients with invasive carcinoma of the cervix, stage Ib, IIa, and IIb proximal. The treatment was based on the association of brachytherapy-colpohysterectomy and lymphadenectomy. Analysis of c-myc gene expression was done using Northern and slot blot hybridization techniques. Overexpression of c-myc (ie, levels at least three times the mean observed in normal tissues) was present in 33% of the tumors. The proportion of carcinomas with c-myc overexpression significantly increased with the size of the primary tumor (P = .04). No relationship was found between c-myc overexpression and the other clinical and histologic parameters, including the nodal status. The relative risk of relapse (overall, pelvic failure, distant metastases) was analyzed in a Cox's proportional hazards model. Three factors were significantly related to the risk of overall relapse when the multivariate analysis was performed, namely, the tumor size, the nodal status, and c-myc expression. A combination of c-myc expression and the nodal status provided a very accurate indication of the risk of relapse. Indeed, patients with negative nodes had a 3-year disease-free survival rate of 93% (95% confidence interval [Cl], 79% to 98%) when c-myc was expressed at a normal level, whereas this rate was only 51% (95% Cl, 26% to 63%) when c-myc was overexpressed (log-rank test, P = .02). In addition, in the subgroup of patients with positive nodes, this rate was 44% (95% Cl, 25% to 77%) and 15% (95% Cl, 4% to 49%) when c-myc gene was expressed at normal level, or overexpressed, respectively. Finally, c-myc gene overexpression was, in the multivariate analysis, the first factor selected by the model regarding the risk of distant metastases.

Postmastectomy radiotherapy: randomized trials.

Postmastectomy radiotherapy decreases threefold the risk of locoregional recurrences according to the results of many randomized trials and overviews. This risk is mainly related to the number of involved axillary nodes (ie, about 25%, 35%, and 55% at 10 years when 1 to 3, 4 to 9, and 10 or more nodes are involved). In contrast, at 10 years, fewer than 15% of patients with negative axillary nodes relapse locally. The effect of postmastectomy radiotherapy on distant metastases and overall survival is a controversial issue. On the one hand, results are compatible with the existence of a mechanism of secondary dissemination generated from locoregional tumor nests. The beneficial effect of radiotherapy may be observed in the absence or presence of adjuvant systemic treatment. On the other hand, a deleterious late toxic, mainly cardiac, effect of radiation has also been shown. This point emphasizes the importance of radiation technique and quality to obtain a positive balance in terms of overall survival.

Analysis of local-regional relapses in patients with early breast cancers treated by excision and radiotherapy: experience of the Institut Gustave-Roussy.

Between 1970 and 1981, 436 patients with T1 and small T2 breast carcinoma were treated by tumor excision followed by radiotherapy at the Institut Gustave-Roussy. The mean follow-up was 5 years, with 50% of patients followed 5 years. Twenty-four patients have experienced a local-regional (LR) relapse for an actuarial LR control rate of 93% at 5 years and 90% at 10 years. Potential prognostic factors for all 24 local-regional recurrences and for the subgroup with relapses in the breast were analyzed. A high Bloom grade and low Nominal Standard Dose (NSD) were significant prognostic factors for predicting LR relapse in both groups. Disease-free survival (from initial presentation) was not adversely affected by a solitary breast recurrence, when patients with successful salvage treatment were considered disease free. However, the group of patients with nodal or dermal recurrences had a much worse prognosis. This paper describes the natural history of breast cancer following a local-regional relapse in irradiated patients without mastectomy. Most importantly, we observed that breast relapses following radiotherapy become clinically apparent more slowly than chest wall failures after mastectomy, and if detected early, that these patients may be successfully retreated.

Long-term effect of internal mammary chain treatment. Results of a multivariate analysis of 1195 patients with operable breast cancer and positive axillary nodes.

A multivariate analysis on 1195 patients with operable breast cancer and histologically positive axillary nodes treated by mastectomy and complete axillary dissection at the Institut Gustave-Roussy between 1958 and 1978 suggests a beneficial effect of treatment of the internal mammary chain (IMC) on the risks of death and distant metastasis for the patients with medial tumors. For these patients, surgical IMC dissection and post-operative irradiation have similar effects on both the risk of death and of distant metastasis. For the patients with lateral tumors, no beneficial effect of the treatment of the IMC on these two risks was observed. Postoperative irradiation to the IMC, axilla, chest wall and supraclavicular nodes significantly decreases the risk of locoregional recurrences independent of the tumor site and surgical management of the lymph nodes.

Ten-year results of a randomized trial comparing a conservative treatment to mastectomy in early breast cancer.

A randomized trial was conducted at the Institut Gustave-Roussy (IGR) between 1972 and 1980 comparing tumorectomy and breast irradiation with modified radical mastectomy. One hundred and seventy-nine patients with an infiltrating breast carcinoma up to 20 mm in diameter at macroscopic examination were included: 88 had conservative management, and 91 a mastectomy. All patients had a low-axillary dissection with immediate histological examination. For the patients with positive axillary nodes, a complete axillary dissection was undertaken. Overall survival, distant metastasis, contralateral breast cancer and locoregional recurrence rates were not significantly different between the two treatment groups.

Latency and time-dependent exposure in a case-control study.

Detailed historical data are elicited often from subjects in retrospective studies, yielding time-dependent measures of exposures. Investigation of a hypothesized period of latency can be made by examining disease/exposure relationships in multiple time windows, either along the age or time-before diagnosis axes. We suggest splitting the data into many time intervals and separately fitting regression models to the available data in each interval. Covariances between estimated coefficients from different intervals are empirically estimated, and used for assessing variability of specified functions of the time-specific coefficients. Alternative methods of interval formation and their consequences are discussed. We apply these methods to a French case-control study of oral contraceptive use and cervical cancer incidence, and compare the results to those of standard analyses.

PIP: A semi-parametric method of fitting regression models to separate time intervals of exposure in retrospective studies is proposed and applied to a set of data on French women with cervical cancer and history of oral contraception. The results were compared to those of several types of conventional analysis. This method is sometimes used for occupational exposure or prospective studies. The data were separated into time intervals by age or time before diagnosis. Covariances between coefficients from different intervals were estimated and used to assess variability of time-specific coefficients. The data were 160 cases, diagnosed with cervical cancer before age 45 at 7 French medical centers from 1982-1985, and 320 matched controls. 88% were from Gustave-Roussy Institute. A logistic regression fit to 4 5-year intervals before diagnosis (right adjusted) produced a trend in coefficients with odds ratios rising from 0.87-1.20, and an approximate relative risk of 2.4. This approach is more suited to detecting a promotor effect of oral contraceptives. Division of the data into 5 5-year age intervals (left adjusted) from age 15 generated a curve with a flatter slope of 1.4, much less significant. In both computations there were stronger positive associations seen with more recent oral contraceptive use. The coefficient for the adolescent age group had a wide variance, suggesting that pill use in those years may be a proxy for multiple partners. The results were consistent with the hypothesis that oral contraceptives act as a tumor promotor with short latent periods, or an accelerator of changes in the cervical epithelium to carcinoma in situ.

Characteristics of reproductive life and risk of breast cancer in a case-control study of young nulliparous women.

Between 1982 and 1985, a case-control study of nulliparous women, aged 25-45, was conducted to analyse the relationships between the risk of breast cancer and causes of nulliparity, including contraceptive methods. Fifty-one cases of breast cancer diagnosed less than 3 months before interview were matched with 95 controls on age at diagnosis, year of interview, and medical center. The causes of nulliparity related to female sterility or subfertility (mechanical or hormonal disorders) were not found to be associated with a significantly higher risk of breast cancer. The causes related to fertilization failure, i.e. no sexual partner, rare sexual intercourse (less than once per month), or partner with abnormal semen, were found to lead to an increased risk. Detailed analysis of contraceptive methods showed that the risk of breast cancer increased (p = 0.02) with a longer duration of use of barrier methods (withdrawal or condom). Conversely, the risk significantly decreased (p = 0.004) with a longer duration of use of non-barrier methods (oral contraceptives, IUD, cap, local spermicides, vaginal douche, safe period, or no method), i.e. methods allowing a direct exposure to human semen.

Risk factors for breast cancer according to age at diagnosis in a French case-control study.

In a French case-control study of 1010 women with breast cancer and 1950 controls with nonmalignant disease, the variations of the effects of 8 risk factors for breast cancer as a function of age at diagnosis, were analysed by tests of homogeneity and trend. The risks associated with a late age at first full-term pregnancy and with nulliparity were different between age-groups (test of homogeneity: p = 0.03), and the highest risks for these two factors were observed in women 45-54 years old. The risks associated with Quetelet index were also found to vary with age at diagnosis (test for trend: p = 0.008). A high Quetelet index decreased the risk of breast cancer in the younger age-groups; this decrease of risk became progressively less important with advancing age, and no such effect was found in the oldest age-group. Inverse results were observed for a tall stature (test for trend: p = 0.04): a tall stature increased the risk of breast cancer in the younger age-groups, and the figures suggested a reverse effect in the oldest group. No large variation with age was found for the effects of age at menarche, history of breast cancer death in mother or sisters, prior biopsy for benign breast disease, and weight. In conclusion, the relative importance of certain risk factors for breast cancer is closely related to age at diagnosis. Nulliparity and a late age at first birth appear to be major risk factors only for middle-aged women, whereas a low Quetelet index and a tall stature appear to increase the risk of breast cancer only for younger women.

Variations in the risk of breast cancer associated with a family history of breast cancer according to age at onset and reproductive factors.

In a case-control study of 495 breast cancer patients and 785 controls between 20 and 56 years of age, the risk of breast cancer associated with a family history of breast cancer was studied according to age and reproductive factors. The familial risk of breast cancer was not significantly modified by age at onset, age at menarche, number of children, age at first full-term pregnancy, menstrual cycle length or age at menopause. However, the familial risk significantly increased with the number of abortions (p < 0.05) and seemed to decrease after a natural menopause (p = 0.08). These results suggest that a familial predisposition to breast cancer exerts the same influence during the first six decades of life, except maybe when there are isolated or repeated events such as abortions or artificially imposed menopause, in which case the risk is apparently greater.

Variation in the interaction between familial and reproductive factors on the risk of breast cancer according to age, menopausal status, and degree of familiality.

BACKGROUND: Studies have found that reproductive factors might have a variable effect on the occurrence of breast cancer (BC) according to the existence or not of a family history of BC. The effect of a family history of BC on the risk of BC may also vary according to the age at diagnosis and the degree of kinship. This may confound the relation between familial risk and reproductive factors. A combined analysis was performed to study the interaction between familial risk and reproductive factors according to degree of familiality, age at interview and menopausal status. METHODS: The present analysis included 2948 cases and 4170 controls in seven case-control studies from four countries. The combined relative risks were estimated using a Bayesian random-effects logistic regression model. RESULTS: The main effects of reproductive life factors on the risk of BC are in agreement with previous studies. Two-way interactions between subject's age or menopausal status and a family history of BC were not significant. Although the three-way interaction between age, familial risk and parity was not significant, familial risk seemed to be increased slightly for women with high parity compared with women with low parity in the older age group, and seemed to be slightly decreased for women with high parity compared with women with low parity in younger women. The subject's age also appeared to have an effect on the interaction between familial risk and the age at first childbirth (P = 0.1). CONCLUSIONS: A possible influence of reproductive and menstrual factors on familial risk of BC has been suggested previously and was also evident in the present study. Three-way interactions between age, family history and parity or age at first childbirth might exist and they merit further investigation.

Oral contraceptive use and risk of benign breast disease in a French case-control study of young women.

We report the results of a French hospital-based case-control study designed to analyse the relation between the use of oral contraceptives (OC) and the risk of benign breast disease (BBD). The cases were 286 women, each less than 46 years old, with BBD histologically verified between 1982 and 1985. Controls were 382 patients, matched to cases on year of birth and month of interview, and who were hospitalized for a non-malignant disease other than BBD. Odds ratios were estimated by multivariate regression, taking into account level of education, place of residence, family history of breast cancer, age at menarche, number of children, age at first full-term pregnancy and Quetelet index. The risk of BBD was found to decrease significantly with a longer use of OC before the first full-term pregnancy (FFTP), but there was no association between the risk of BBD and the duration of OC use after FFTP. OC use before FFTP reduced the risk of non-proliferative disease, but did not significantly affect the risk of proliferative disease. These results did not depend on the amount of oestrogen (0.05 mg or more vs < 0.05 mg) contained in OC.

PIP: This article reports the findings of a French-based case-control study examining the relationship between risk of benign breast disease (BBD) and oral contraceptive (OC) use. Special emphasis was given to duration of OC use before and after the first full-term pregnancy (FFTP). Conditional multivariate logistic regression was the statistical process used to analyze the findings. 286 women with verified BBD were studied. 382 women, who were hospitalized for a non-BBD disease, were used as controls. Each case was matched with at least one control. 7 confounders were adjusted for. Confounders included family history of breast cancer, level of education, place of residence, age at menarche, number of children, age at FFTP, and Quetelet index. The risk of BBD decreased with longer OC use before the FFTP. There was no association observed between risk of developing BBD and the duration of OC use after the FFTP. OC use before the FFTP reduced the risk of non-proliferative disease, but did not significantly affect the risk of proliferative disease.

Is it useful to remove internal mammary nodes in operable breast cancer?

From September 1963 to January 1968, 243 patients with operable breast cancer were included in a randomized trial designed to compare classical radical mastectomy (RM) alone to extended mastectomy (EM), i.e. RM plus internal mammary dissection. One hundred and seventeen patients underwent RM, and 126 EM. After a mean follow-up time of 20 years, no significant differences were observed between the two treatment groups for overall survival, for the relapse-free survival rates, nor for distant metastasis, or locoregional recurrence rates. From a regression model, significant interactions were found between risk of death, EM, and both nodal status and site of the tumour. When compared to RM, EM significantly decreased the risk of death for patients with internal or medial tumour and positive axillary nodes (P = 0.05). No beneficial effect of EM was observed for any of the other patients; on the contrary, EM seemed to increase the risk of death for the patients with external tumour and negative axillary nodes (P = 0.07).

The c-myc proto-oncogene in invasive carcinomas of the uterine cervix: clinical relevance of overexpression in early stages of the cancer.

The biological behavior of early-stage invasive carcinoma of the uterine cervix is not always predictable. Therefore it is important to identify new biological markers which could more accurately predict the evolution of the disease. Amplification and/or overexpression of the c-myc gene were frequently observed in advanced-stage cervical cancers and were shown to be associated with tumor progression. More interesting was the study on 93 patients with early-stage carcinoma showing that c-myc gene overexpression was significantly related to a higher risk of relapse. A combination of c-myc expression and nodal status provided a very accurate indication of the risk of relapse. Indeed, in the subgroup of patients with negative nodes, the 3-year disease-free survival rate was 93% (95% confidence interval CI: 79-98%) when c-myc was expressed at a normal level, whereas this rate was only 51% (95% CI: 26-63%) when c-myc was overexpressed. Moreover the c-myc overexpression was related to a 6.1-times higher risk of distant metastases, suggesting that activation of this proto-oncogene may lead to metastatic ability of tumor cells. These data clearly show that patients with c-myc overexpression are high risk patients who thus might benefit from intensive treatment.

Can breast reconstruction with gel-filled silicone implants increase the risk of death and second primary cancer in patients treated by mastectomy for breast cancer?

An increased risk of cancer and autoimmune disease associated with gel-filled silicone implants has been suggested recently, but these possible detrimental effects have not been adequately studied in patients with breast cancer. In order to evaluate these effects, we have studied 146 patients with breast cancer treated by mastectomy at the Gustave Roussy Cancer Institute between 1965 and 1983 and who received a gel-filled silicone implant for immediate or delayed breast reconstruction between 1976 and 1984. These patients were compared with 146 matched controls with breast cancer who were treated in the same center by mastectomy without breast reconstruction and were matched for age at diagnosis (within 10 years), year of diagnosis (within 3 years), stage, histologic type of the tumor, histopathologic grade, and nodal status. The relative risks of death, relapse, and second primary cancer were estimated by means of the Cox proportional hazards model stratified on age at diagnosis. The risks of distant metastasis and death due to breast cancer were significantly lower in the breast reconstruction group than in the control group. The risks of local recurrence, second breast cancer, and second primary cancer in another site than the breast were not significantly different between the two groups of patients. Our results do not support the hypothesis of a detrimental effect of gel-filled silicone implants either in the course of breast cancer or in the risk of death due to other diseases.

[Oral contraception and benign breast diseases: point of view of the epidemiologist]. / Contraception orale et mastopathies bénignes: le point de vue de l'épidémiologiste.

The study of the association between oral contraceptives (OC) and benign breast disease (BBD) is of great interest for the epidemiologist, because of the impact that this association may have on the risk of breast cancer. In fact, three factors have to be analysed simultaneously: OC use, the occurrence of BBD and the occurrence of breast cancer. This analysis is complex because two out of the three factors are not univocal in their definition (various composition and doses for pills, numerous histologic types of BBD). Moreover, a fourth parameter is to take into account: the time factor. This last factor intervenes in the chronological order of events (OC before or after BBD), and for the use of different types of pills (duration of use, age at the start of OC use before or after the first full-term pregnancy, etc). Although all the relationships between these different factors are still insufficiently known at the present time, certain points are acquired: 1) OC use decreases the risk of BBD (particularly when pills contain high doses of progestogens); 2) a history of BBD increases the risk of breast cancer (particularly when BBD is a fibrocystic mastosis with cellular atypia); 3) in spite of the decrease in the risk of BBD associated with OC use, OC use does not decrease the risk of breast cancer. For a long time, this third proposition appeared as paradoxal. In 1986, Stadel and Schlesselman tried to estimate the terms of this paradoxe. Their estimations allowed to show that the risk of breast cancer associated with OC use, for women with a history of BBD, was overestimated in all previous reports. Indeed, the decrease of the occurrence of BBD due to OC use, which was associated with an important reduction in the number of breast cancer cases, was not taken into account in the estimation of this risk. Other epidemiological investigations with a greater number of subjects are needed to further analyse complex relationships between OC use, BBD and breast cancer.

[Pattern of mortality from ovarian cancer in France 1968-1991]. / Evolution de la mortalité par cancer de l'ovaire en France entre 1968 et 1991.

In a recent meta-analysis of three American case-control studies, Whittermore et al found an increased risk of ovarian cancer in women treated for infertility. We have studied the mortality due to ovarian cancer in France between 1968 and 1991 in order to verify whether the use of these treatments could have modified mortality rates during the 24 years period studied. First, we have noted that among the 3175 deaths from ovarian cancer registered in France in 1991, only 136 occurred in women under 45 years of age, and 943 occurred in women between 45 and 59 years of age. We then observed an important geographical diversity with a higher frequency of ovarian cancer in the North of France as compared to the South. Finally, we observed a 72% increase in mortality during the study period. This increase can be entirely explained by the increase in the mortality rates among women over 60 years of age. Among women under 45, the ovarian cancer rate decreased by 24% between 1968 and 1991. Among women between 45 and 59 years of age, these rates increased by only 8% (or about 0.3% per year). These results imply that a possible effect--in terms of public health--of treatment for infertility does not necessarily translate into a significant increase in mortality from ovarian cancer among women under 60 years of age. This result in no way nullifies the role that epidemiological studies have in verifying the accuracy of risks associated with treatments for infertility.

[Factors linked to the development of pleuro-pulmonary metastases in breast cancer. Preliminary study]. / Facteurs liés à l'évolution des métastases pleuro-pulmonaires dans le cancer du sein. Etude préliminaire.

The risk of death after pleuro-pulmonary metastasis was studied on a sample of 55 patients treated for breast cancer between 1954 and 1978. The women treated by tamoxifen, aminoglutethimide, or radiotherapeutic castration, at the time of the appearance of their metastasis, had a significantly lower risk of death than the women treated by other methods, or who were left untreated. The characteristics of the initial tumor, the delay between the primary tumor and the metastasis, the rhythm of systematic chest X-rays, and the use of chemotherapy did not significantly decrease this risk. The management of controlled therapeutic trials is suggested to confirm these results.