转移性乳腺癌预后生存影响因素：一项单中心研究经验

 

Prognostic factors and survival in metastatic breast cancer: A single institution experience

• Afef Khanfir^a, 
• Faiez Lahiani^a, 
• Racem Bouzguenda^a, , , , 
• Ines Ayedi^a, 
• Jamel Daoud^b, 
• Mounir Frikha^a

• ^a Department of Medical Oncology, Habib Bourguiba University Hospital, 3029 Sfax, Tunisia
• ^b Department of Radiotherapy, Habib Bourguiba University Hospital, 3029 Sfax, Tunisia

• http://dx.doi.org/10.1016/j.rpor.2013.01.001, How to Cite or Link Using DOI

• Permissions & Reprints

---

Keywords

• Breast cancer; 
• Metastasis; 
• Prognostic factors; 
• Survival

---

1. Background

Breast cancer is the most common malignant disease among women in the world.¹ Despite the advances in the diagnosis and in the treatment of breast cancer, 6–10% of affected patients present metastatic breast cancer at diagnosis and 30–40% will develop metastasis during the evolution of their disease.^2 and 3 The peak of relapses is seen between 2 and 3 years.² The onset of metastases, as a sign of incurability, constitutes a major problem of care.⁴

The emergence of new drugs and recent therapeutic strategies have not only helped to extend the median survival time (around 30 months now) but also to improve the quality of patients’ lives.⁵

The aim of this retrospective study is to identify the prognostic factors influencing the survival of patients affected with metastatic breast cancer taken over in the south of Tunisia.

2. Materials and methods

This study included 332 patients with histologically proven breast cancer with synchronous (SM) or metachronous (MM) metastasis (MTS). The study was conducted over eight years (from January 2000 to December 2007). Data were updated on January 2010. We collected, for each patient, the epidemiological and clinical data, the therapeutic modalities and the evolution. The treatment of the primary tumor consisted, according to the operability and the response to chemotherapy (CT), in the surgery (mastectomy or tumorectomy with or without axillary lymph nodes dissection) and/or radiotherapy (RT) at a dose of 36 Gy in 12 sessions. The metastasectomy and/or the RT had been indicated for solitary metastasis. The RT had been, also, used for painful bone metastasis, in case of spinal cord compression, for consolidation of threatening bone lesions and/or in case of cerebral metastasis with a dose of 30 Gy in 10 sessions. The first line systemic treatment for metastatic relapse consisting in hormone therapy (HT) or chemotherapy (CT) depends on the disease free interval (DFI) between the end of the adjuvant treatment and the metastatic relapse, of the positivity of hormonal receptors (HR), characteristics of metastasis and the patient PS based on the WHO classification. Bisphosphonates (Pamidronate, Zoledronate or Ibandronate) were prescribed in cases of bone metastases.

The OS was defined as the time between the date of diagnosis and the date of death or last visit for the group of patients presenting SM and as the time between the date of relapse and the date of death or last visit for patients with MM using the Kaplan–Meier. The OS was analyzed in relation to different prognostic factors: the age (≤50 years vs. >50; ≤70 years vs. >70; ≤35 years vs. >35), the menopausal status (premenopausal vs. postmenopausal); the presentation of metastatic disease (SM vs. MM), the disease free interval (DFI) (≤24 months vs. >24, ≤60 months vs. >60), the performance status (PS) (0–1 vs. >1), the estrogen receptor (ER) (ER+ vs. ER−); the hormone receptors for progesterone (PR) (PR+ vs. PR); the number of metastatic sites (1 site vs. >1 site); the nature of the metastatic site (visceral vs. non visceral), the type of the first line medical treatment (CT vs. HT), the surgery of the primitive tumor (SPT) vs. no SPT, the loco regional RT (LRRT) vs. no LRRT and the use of biphosphonates (biphosphonates vs. no biphosphonates). The study of survival based on prognostic factors was performed using the log-rank test for univariate analysis and Cox model for multivariate analysis.

3. Results

3.1. Characteristics of patients and of the disease

There were 129 cases of SM (39%) and 203 cases of MM (61%). The median age was 50.5 years (range: 25–85 years). The PS was 0–1 in 72% of cases. Fifty-five percent of our patients were premenopaused at diagnosis. The anatomopathological examination concluded to an invasive ductal carcinoma in 82% of the cases, a II or III histological grade in 98% of cases and a positive HRs in 56% of the cases. Her2 status was determined by immunohistochemistry for 3 patients, this marker was over expressed in one of them. The rate of CA15-3 was increased in 72% of the specified cases. The median time of relapse in MM cases was 31 months (2–248 months). The distribution of the metastatic locations is detailed in Table 1. One hundred and fifty-seven patients (47.3%) had a single metastatic site.

3.2. Treatment

3.2.1. Surgery

Two hundred and ninety-eight patients were treated. Fifty-two patients were treated by surgery for their primitive tumor: 39 underwent a mastectomy with axillary node dissection, eleven had a mastectomy without axillary node dissection and two had a lumpectomy with axillary node dissection. Twelve patients had surgeries of secondary localizations: Hepatic metastasectomy (two patients), cerebral metastasectomy (two patients), bilateral annexectomy (one patient), spinal lymph node dissection (one patient) and bone consolidating surgery (six patients).

3.2.2. Medical treatment

Amongst the 298 treated patients, 283 patients (94.6%) received a first line medical treatment consisting in CT (258 patients) or HT (25 patients). The untreated patients were lost to follow up at diagnosis of metastases or had a poor PS preventing treatment. Among the 265 patients who received CT (in first line or after first line HT), 119 (44.9%) were treated with second line CT and 47 (17.7%) with 3 lines of CT or more.

The first line CT was anthracycline-based in 137 cases (51.6%), taxane-based in 88 cases (33.2%), vinorelbine-based in 48 cases (18.2%) and platinum-based in 19 cases (7.3%). Fifteen patients had a combination of anthracycline with taxane according to a sequential schedule in 5 cases (1.9%) and concomitant schedule in 8 cases (3%).

After receiving the first line CT, the objective response rate was 65% and the median time for progression was 9 months.

One hundred and eight patients were treated with HT (immediately to the diagnosis of the metastatic disease in 25 cases, a relay after a response to the first line CT in 73 cases or after the failure of a first line CT in 10 cases). Forty-one percent of these patients underwent second-line HT and thirteen percent had at least three lines.

Out of 225 patients with bone metastases, only 94 patients (41.7%) received bisphosphonates. Pamidronate was the most widely used molecule. The average duration of the prescription of bisphosphonates was 18.6 months (1–96 months).

3.2.3. Radiotherapy

One hundred and fifty-five patients underwent RT, 15 of them had only a palliative RT without systemic treatment. It was brain RT (54 cases), bone RT (78 cases) and/or locoregional RT (23 cases). The response (tumor size reduction) to the LRRT was observed in 67% of cases. The response to the palliative RT was variable depending on location. Seventy-one percent of patients with brain metastases showed a disappearance of intracranial hypertension symptoms after RT. The relief of the spinal cord compression was observed in 69% of irradiated patients. The effectiveness of RT for pain management was observed in 90% of cases. The average duration of analgesic response to radiotherapy was 26 weeks.

3.3. Survival and prognostic factors

In our study, the OS was respectively 63%, 42%, 26% and 12% at 1, 2, 3 and 5 years (Fig. 1).

The median survival of patients who received treatments was 18 months. It was 23.3 months, 29.3 months, 18 months, 15 months and 11 months, respectively, in case of only: bone, lymph node, hepatic, lung or brain MTS.

In patients with single bone MTS, the 5 years OS was 36% and it was 23% for those with only lymph node metastasis.

The median survival for patients who underwent metastasis resection was 83 months.

3.3.1. Univariate analysis

Using univariate analysis, the following prognostic factors were found to influence positively the survival: Age ≤ 70 years, good PS (0–1), positive ER, positive PR, single metastatic site, non visceral MTS, using HT in first line metastatic treatment, locoregional treatment (RT or/and surgery) in case of SM and the use of bisphosphonates (Table 2).

3.3.2. Multivariate analysis

In case of multivariate analysis, improved survival was associated significantly with good PS (1–2), non visceral MTS and age ≤ 70 years (Table 3).

4. Discussion

In our study, the median survival of SM and MM breast cancer patients who received treatment was 18 months and the 5 years OS was 12%. Our results were similar to those reported in many previous studies in which the median survival ranged between 16 and 34 months and the 5 years OS varied from 4% to 28%.^{6, 7,8, 9 and 10}

For many authors, age at diagnosis is acknowledged as a prognostic factor affecting the survival. Dawood et al. did not find age to be a prognostic factor for predicting survival.² However, according to the study of Largillier et al., poor prognosis was reported with age over 50 years.¹¹ In our material, we studied survival according to differing age groups (≤ or >35 years); (≤ or >50 years) and (≤ or >70 years). We identify that the age of more than 70 years was a pejorative factor influencing significantly the survival in univariate and multivariate analysis.

Some studies have shown that premenopausal women have a better prognosis than postmenopausal patients.² This finding was not confirmed in our study and the influence of the menopausal status on survival was not significant.

Several studies have shown that the more altered the PS is, the worst the prognosis.^{12, 13, 14 and 15} Our study confirmed this finding in univariate and multivariate analysis.

Many previous studies have shown that the DFI is an important prognostic factor.^{2, 14, 16, 17, 18, 19, 20 and 21}Vogel et al. reported in a series of 433 patients with recurrent breast cancer a median survival from first relapse of 26 months for patients with DFI < 24 months and 44 months for patients with DFI > 24 months (p < 0.01).²⁰ Our study confirmed these findings; relapse after 2 years of adjuvant therapy was a good prognostic factor.

Dawood et al. found that the median survival in women with SM was 12 months longer than women with MM, the difference was statistically significant in univariate and multivariate analysis.² Our study did not confirm these data and the difference in survival between these two groups was not statistically significant.

The non visceral localizations (skin, soft tissue and bone) have a better prognosis than visceral sites. The most pejorative visceral sites are the brain, liver and lung.^12 and 21 Our study confirmed these results as well in univariate or multivariate analysis.

In our study, the 5 years OS and the median survival with single bone MTS were respectively 36% and 23.3 months. In other previous studies, median survival ranged between 24 and 48 months and the 5 years OS achieved 20%.^{22, 23 and 24} In a study conducted by Coleman and Rubens concerning patients whose cancers were diagnosed in the 1970s and 1980s, the median survival was 24 months in patients with first recurrence in the skeleton compared with 3 months after first relapse in the liver (p < 0.01).²⁵

Hepatic metastases are generally associated with a poor prognosis. Pentheroudakis et al. reported in a retrospective study of 500 liver metastasis breast cancer cases a median survival of 16.4 months and 5 years OS of 8.5%.²⁶ Reported median survivals in a phase III EORTC trial which investigated the first line CT in metastatic breast cancer were respectively 27.1 months and 16.8 months in patients with liver MTS alone and with liver with other sites of MTS.⁴ In our study, median survival was 17.7 months in cases of hepatic MTS only. Long survival (83 months) was obtained in one patient who underwent surgical resection of liver MTS.

The median survival for patients with only lung MTS ranges from 6 to 15 months as noted in previous studies.^24 and 27 In our study it was 14.6 months.

The majority of patients with cerebral MTS died in 40% of cases because of non control of the latter.²⁸ The median survival depends on the application of treatment: if not treated, this median is 1 month; and 2 months for patients who had received corticosteroids only, while it grows to 3–6 months if the whole brain RT is carried out.²⁸ In our study this median was 11.2 months.

The median survival in case of ovarian and/or peritoneal metastases, according to previous studies, ranges from 6.4 months to 23 months. However, long term survival reaching more than 12 years has been reported particularly in cases of metastatic breast cancer with positive hormonal receptors.^29 and 30

Many retrospective recent studies have shown that the surgical treatment of the primitive tumor improve the OS rate for patients presenting breast cancer with SM.^{31, 32 and 33} This was confirmed by our study.

The Le Scodan et al.³³ retrospectively reviewed a mono centric study of 581 treated patients for an SM breast cancer. Of that number, 320-patients had a loco regional treatment (exclusive RT for 78% of them, a surgery followed by a RT for 13% and exclusive surgery for 9%), in addition to systemic treatment against 261 patients who received systemic treatment only. After a median follow-up of 39 months, the 3-year OS was 43.4% for patients who underwent a local therapy, against 26.7% for patients who had only systemic treatment (p < 0.02). Our study showed the benefit provided by a local therapy (RT and/or surgery) on OS for patients with synchronous metastases of breast cancer.

The use of biphosphonates in case of bone MTS reduced the skeletal-related events (SREs) rate, delayed the time to SREs and prolonged the time in developing bone pain, but it do not appear to affect survival.³⁴However, in our study we noticed that the use of biphosphonates improves the overall survival.

5. Conclusion

Our therapeutic results in terms of survival were correlated with those of previous studies. Our study confirmed the positive impact, influencing independently the duration of survival in metastatic breast cancer, of the good general state, of the age ≤ 70 years, and of the absence of visceral metastasis.

Conflict of interest

No conflict of interest.

Financial disclosure

Not applicable.