Factors Associated with the Selection of First-line Bevacizumab plus Chemotherapy and Clinical Response in HER2-negative Metastatic Breast Cancer: ONCOSUR AVALOX Study

Discussion

The AVALOX study showed that bevacizumab was mostly combined with single-agent taxane-based chemotherapy, mainly paclitaxel, in patients with HER2-negative MBC, including those with TNBC and HRPBC. Hence, we identified factors involved in the selection of bevacizumab plus paclitaxel, as the most frequent first-line regimen used in clinical practice in Spain. The findings of this analysis suggest that the number of metastatic sites (<3 vs. ≥3 sites) and ER status were factors significantly associated with the choice of first-line treatment with bevacizumab plus paclitaxel, although ER-positive status was only identified as an independent factor associated with the selection of this approach. However, the statistical significance of differences may be explained by the large number of patients with ER-positive MBC included in this series.

Thus, bevacizumab plus paclitaxel seems to be used regardless of other clinical risk factors in HER2-negative MBC, such as location of metastases and ECOG PS (16), and this regimen seems more likely to be used in patients with multiple metastases (≥3 sites), although this clinical factor did not reach statistical significance in the multivariate model.

Of note, bevacizumab plus paclitaxel was the most common regimen used among patients with TNBC, with more than half of this sub-group receiving this regimen, which seems to reflect an increasing use of bevacizumab and single-agent taxane in this sub-group of patients with few treatment options and typically associated with a more aggressive disease and a poor prognosis.

Our findings also suggest that bevacizumab plus chemotherapy treatment, mainly based on single-agent paclitaxel, is an effective treatment option in patients with HER2-negative MBC. As a descriptive comparison only, the efficacy figures in terms of ORR (63.7%) are within the range reported in the phase III clinical trials with bevacizumab and taxane-based chemotherapy (3, 4) where an ORR ranging from 37% to 64% has been shown. In addition, the bevacizumab plus chemotherapy strategy was also effective in TNBC. It is noteworthy that the benefit obtained in patients with TNBC seems to be similar to that observed in patients with HRPBC disease. Similarly, a recent analysis of the ATHENA trial also showed comparable OS between patients with HRPBC and TNBC when adjusted for prognostic factors (16). The AVAREG study, where half of patients had TNBC, has recently confirmed that first-line bevacizumab plus paclitaxel is an effective and well-tolerated treatment option for patients with HER2-negative MBC (18).

In our series, an age between 36-50 years, metastases at other sites than bone and the presence of multiple metastases (≥3 sites) were identified as independent factors significantly associated with the achievement of objective responses to bevacizumab plus chemotherapy, mainly based on paclitaxel, in the overall study population. Similarly, patients aged 36-50 years were also more likely to achieve an objective response in the HRPBC subgroup, while metastases at other sites than bone was the only factor associated with response in patients with TNBC patients.

The potential prognostic value of age remains uncertain but evidence from the literature points to a worse survival at older ages. Thus, a previous Spanish series reported significant differences in survival at two years according to age in patients with HER2-negative MBC, showing a significantly higher proportion of patients alive at 2 years among the younger population (<35 years and 36-50 years) (14). In our series, we identified an age between 36-50 years old as a predictive factor of clinical response. However, it is noteworthy that age at diagnosis does not seem to influence first-line treatment selection and the majority of patients older than 70 years (77%) received bevacizumab and paclitaxel combination therapy in our series.

Concerning the predictive and prognostic role of metastatic location, when disease is confined to bone and soft tissue, there is generally a more indolent course and survival is longer, while patients with brain, liver and lung metastases usually have a more aggressive condition and a worse outcome. In addition to metastasis location, the number of metastasis sites has been identified as an important prognostic factor in patients receiving first-line bevacizumab plus chemotherapy for HER-2 negative MBC, with ≥3 involved organs associated with a worse outcome (16). Our results suggest that patients with metastases at other sites than bone are more likely to achieve objective responses than those carrying bone metastases, who also had metastases in the liver (52%) and/or lung (39%), typically associated with a worse outcome. Accordingly, a previous study of patients with MBC showed that the median OS dropped from 2.6 years when bone was the only location of disease to one year if other organs were involved (19).

All findings together suggest that patients harboring clinical factors typically associated with more aggressive disease, such as multiple metastatic sites, with ≥3 organs involved, and metastases at other sites than bone, respond well to bevacizumab plus chemotherapy, mainly single-agent taxane. Therefore, this combination may offer an active first-line treatment option for patients usually associated with a clinically defined aggressive disease and poor outcome. Accordingly, a Spanish group of oncology experts has recently recommended the addition of bevacizumab to standard chemotherapy as first-line treatment for those patients with clinically aggressive disease (20).

In addition, other factors previously associated with a poor prognosis in HER2-negative MBC, such as a DFI <12 months, prior therapy with neo (adjuvant) taxanes and TNBC (16), do not seem to influence the achievement of objective responses in our series. Accordingly, a meta-analysis of the E2100, AVADO and RIBBON-1 trials showed that the benefit of adding bevacizumab to chemotherapy was maintained across all patient subgroups regardless of DFI, prior adjuvant taxane use, the number of metastatic sites or visceral metastases (21). In addition, we found that patients with high-risk HRPBC, defined as those presenting with at least two risk factors (DFI <12 months, prior (neo) adjuvant taxanes and/or ≥3 metastatic sites or liver metastases), had a high response rate (>70%). The ATHENA study, which defined similar factors of poor prognosis, showed that patients with HRPBC and two or more risk factors had a particularly poor prognosis in terms of OS, even shorter than those patients with TNBC (16).

The findings from our analyses also suggest that bevacizumab plus chemotherapy treatment, mainly containing paclitaxel, is well-tolerated in patients with HER2-negative MBC. Bevacizumab-related toxicity was generally mild and manageable in our series. Only 2% of patients experienced grade 3 toxicities and no grade 4 toxicities were detected. In addition, no bevacizumab-related deaths were reported. Safety findings were consistent with phase III trials with bevacizumab and taxanes, with no new or unexpected signals detected (22).

The addition of bevacizumab to paclitaxel has shown no significant adverse effects that interfere with HRQL (23). The results from the E2100 study found that the addition of bevacizumab to paclitaxel was not associated with additional side effect burden from the patient perspective and was related with a greater reduction in breast cancer-specific concerns (24). Our series seems to reflect this trend, and it is interesting to note that FACT-B scores were similar between the overall population, where three-quarters of patients had HRPBC, and the subgroup of patients with TNBC. This finding may suggest that first-line treatment with bevacizumab and taxane-based chemotherapy may be adequate from the patient's perception and its impact on quality of life even in those patients with TNBC, usually associated with an adverse outcome.

In addition to the obvious limitations arising from the observational nature of this study, other limitations should be considered. Firstly, the cross-sectional design of the study does not allow a follow-up period to collect long-term survival data. Therefore, the prognostic value of the clinical factors related to the achievement of response could not be confirmed. Nevertheless, this was not the objective of the study but to reflect the factors that may influence treatment decision-making and identify clinical predictive factors of response. Finally, considering the lack of a follow-up period, the authors acknowledge that a certain degree of underreporting of mild adverse events could have occurred.

Despite the above limitations, to our knowledge, our series of nearly 300 patients with HER2-negative MBC is the largest series where the factors involved in the selection of bevacixumab and paclitaxel and clinical predictive factors for response have been identified in the context of routine oncology practice in Spain. Moreover, our findings, although modest, might provide a welcome addition to the scarce data on predictive factors for efficacy, and given the lack of response biomarkers, help identify patients with HER2-negative MBC who would derive most benefit from the addition of bevacizumab to single-agent taxane-based chemotherapy in clinical practice.

In summary, an ER-positive status appears to be the only independent factor associated with choosing this approach in HER2-negative MBC, although this finding may be influenced by the high proportion of patients with ER-positive MBC in this series. Therefore, we have not found a clear clinical profile for patients with MBC in using bevacizumab and single-agent taxane in clinical practice in Spain. However, our findings suggest that aggressiveness of disease rather than age and other biological factors seem to influence the selection of bevacizumab and paclitaxel in HER2-negative MBC. In addition, first-line bevacizumab plus chemotherapy, mainly based on paclitaxel, seem to be notably more effective in patients who harbor more aggressive disease, defined by large metastatic burden (≥3 sites) and metastases at other sites than the bone. Further studies are required to assess the prognostic implications of these clinical factors that may allow for identification of patients who could derive most benefit from first-line bevacizumab and single-agent taxane-based chemotherapy in patients with HER2-negative MBC in routine oncology practice.