Survival no better after primary tumor removal in metastatic breast cancer

SAN ANTONIO – Surgical removal of the primary tumor and affected lymph nodes afforded no overall survival benefit in women who had metastatic breast cancer at initial presentation, according to a pair of randomized trials presented at the San Antonio Breast Cancer Symposium.

In a first trial, conducted among 350 women in India who had responded to initial chemotherapy, about 20% of the women were still alive at 5 years regardless of whether they had surgery or just received more systemic therapy, first author Dr. Rajendra Badwe, director of the Tata Memorial Hospital in Mumbai, India, reported in a session and at a press briefing.

Superior locoregional control conferred by surgery was canceled out by a higher risk of progression in distant sites, lending support to more than 20-year-old preclinical data by Dr. Bernard Fisher and his colleagues suggesting that an intact primary suppresses growth of distant metastases (Cancer Res. 1989;49:1996-2001).

"Locoregional treatment of the primary tumor in women presenting with metastatic breast cancer did not result in any overall survival benefit and hence should not be offered as a routine practice," Dr. Badwe commented.

"The biological fallout of this study is that surgical removal of the primary tumor in these women appears to confer a growth advantage on distant metastases. ... This is the first time that we have evidence in human studies that locoregional treatment has a great kinetic effect on distant metastases," he added.

In a second trial, conducted among 293 women in Turkey with untreated de novo metastatic breast cancer, median estimated survival was statistically indistinguishable, at about 3.5 years, regardless of whether women had up-front surgery or simply systemic therapy, first author Dr. Atilla Soran reported in a session on behalf of the Turkish Federation of Societies for Breast Diseases.

Subgroup analyses suggested that surgery conferred a survival advantage among women with solitary bone metastases but a survival disadvantage among women with multiple liver or lung metastases.

The rate of locoregional progression was one-fifth as high with surgery as with systemic therapy.

"There was no statistically significant difference in overall survival at early follow-up, but we need longer follow-up," Dr. Soran commented. "There were potentially important subgroup differences."

The two trials have implications – both for other ongoing trials and for patient care – according to invited discussant Dr. Seema A. Khan, a professor of surgery and Bluhm Family Professor of Cancer Research at Northwestern University, Chicago, and a physician at the university’s Lynn Sage Breast Center.

"A large benefit of primary site local therapy seems unlikely based on the data we saw today," she maintained. "The assumptions we have used in our ongoing trial designs will need to be reassessed. Preplanned pooled analyses may yield sufficient power to detect smaller differences."

As for patient care, "it is pretty clear that at this point in time, we have to make sure that our patients understand that there is really no proven survival advantage to primary site local therapy," Dr. Khan said. "So I don’t think this treatment should be offered to patients with asymptomatic tumors unless they are participating in a clinical trial. There may be a local control advantage; we need to see more data on that."

Indian trial

Participants in the Indian trial, conducted between 2005 and 2013, were women with de novo metastatic breast cancer who had had a complete or partial response to anthracycline chemotherapy alone or with a taxane.

They were randomized evenly to receive locoregional therapy (lumpectomy or mastectomy with axillary lymph node dissection, plus radiation therapy to the chest wall or breast and lymph nodes) or no locoregional therapy as a control. Both groups received hormonal therapy if indicated.

In the control group, about 10% of women underwent a palliative mastectomy because of impending fungation or pain in the breast, as permitted by study protocol, according to Dr. Badwe.

The patients thus received contemporary therapy, he said, except that the 16% with HER2-positive disease did not receive the targeted agent trastuzumab (Herceptin).

Results showed that median overall survival was about 18 months, and the 5-year rate was 19.2% with locoregional therapy and 20.5% without it, a nonsignificant difference. "Uniformly, there was no difference at all in any subsets," Dr. Badwe reported.

The study had a one-sided superiority design, he noted. "We wouldn’t have been able to see a 2.5- or 3-month difference, or a 4% detriment" in overall survival.

The surgery group had a lower risk of local progression-free survival events (hazard ratio, 0.16; P = .00), but a higher risk of distant progression-free survival events (HR, 1.42; P = .01).

Several theories might explain why removal of the primary would accelerate growth of metastases, according to Dr. Badwe.

"The first and the foremost is that the act of surgery itself might elaborate some growth factors which might allow metastatic disease to grow. The second possibility, which was suggested by Dr. Fisher (Cancer Res. 1989;49:1996-2001) is that the primary tumor, which predates the onset of distant metastases, elaborates some inhibitory factors, and they are not there once the primary tumor is removed, bestowing autonomy of growth on the distant metastases," he explained. "And the third possibility is the act of surgery might induce some more metastatic processes by dissemination and create new disease."

Session attendee Dr. Steven Vogl, an oncologist in the Bronx, N.Y., said, "The ascertainment of disease progression requires disease that you can follow. A woman with bone-only metastases with her cancer in place, you can tell when her cancer is getting worse because the primary site or the axillary nodes are getting bigger. It’s much more difficult if you’ve taken those off and irradiated the chest wall. Have you looked at your data to see if that’s what was going on, why you had more distant metastases, because they couldn’t progress locally? This is a medical trial explanation that contradicts Fisher’s biologic hypothesis."

"There was a fixed time duration at which systemic investigations were performed to assess whether the distant metastases progressed," Dr. Badwe replied. If anything, the patients who did not have surgery had more assessments of their distant metastases, he said.

Dr. Tari A. King, a session attendee from the Memorial Sloan-Kettering Cancer Center, New York, noted that a lack of HER2 therapy in the trial may have had a large effect.

"We do have prospective registry data here [abstract 18-09, presented in a poster session] from a trial that we completed in the United States sponsored by the Translational Breast Cancer Research Consortium, and the patients in our study, whether they received surgery or not, their 2-year overall survival is far superior to what you’ve just showed us," she commented. "So I’m not sure that we can really apply your data to the modern targeted therapy regimens that we see in the United States."

Turkish trial

The Turkish trial, known as the MF07-01 trial, was conduced between 2008 and 2012 among treatment-naïve patients.

They were randomized evenly to receive either systemic therapy alone or surgery for the primary tumor (with or without axillary dissection) followed by radiation therapy if indicated, plus systemic therapy.

All patients received hormonal therapy as needed, and those with HER2-positive disease received trastuzumab.

With a median follow-up of 18 months, the median overall survival was 46 months with initial surgery and 42 months with initial systemic therapy, a nonsignificant difference, reported Dr. Soran, who disclosed no relevant conflicts of interest.

In unplanned subgroup analyses, the findings were similar for most subgroups of patients. However, surgery yielded superior survival in patients with solitary bone metastases (not reached vs. 42 months, P = .02) and inferior survival in patients having multiple liver or pulmonary metastases (16 months vs. not reached, P = .02).

The rate of locoregional progression was much lower with initial surgery than with initial systemic therapy (0.7% vs. 3.6%).

Dr. Soran emphasized that the trial’s planned median follow-up is 36 months, so the presented results are only preliminary. Quality of life and morbidity analyses are ongoing.

Dr. Badwe and Dr. Soran disclosed no relevant conflicts of interest.

SAN ANTONIO – Surgical removal of the primary tumor and affected lymph nodes afforded no overall survival benefit in women who had metastatic breast cancer at initial presentation, according to a pair of randomized trials presented at the San Antonio Breast Cancer Symposium.

In a first trial, conducted among 350 women in India who had responded to initial chemotherapy, about 20% of the women were still alive at 5 years regardless of whether they had surgery or just received more systemic therapy, first author Dr. Rajendra Badwe, director of the Tata Memorial Hospital in Mumbai, India, reported in a session and at a press briefing.

Superior locoregional control conferred by surgery was canceled out by a higher risk of progression in distant sites, lending support to more than 20-year-old preclinical data by Dr. Bernard Fisher and his colleagues suggesting that an intact primary suppresses growth of distant metastases (Cancer Res. 1989;49:1996-2001).

"Locoregional treatment of the primary tumor in women presenting with metastatic breast cancer did not result in any overall survival benefit and hence should not be offered as a routine practice," Dr. Badwe commented.

"The biological fallout of this study is that surgical removal of the primary tumor in these women appears to confer a growth advantage on distant metastases. ... This is the first time that we have evidence in human studies that locoregional treatment has a great kinetic effect on distant metastases," he added.

In a second trial, conducted among 293 women in Turkey with untreated de novo metastatic breast cancer, median estimated survival was statistically indistinguishable, at about 3.5 years, regardless of whether women had up-front surgery or simply systemic therapy, first author Dr. Atilla Soran reported in a session on behalf of the Turkish Federation of Societies for Breast Diseases.

Subgroup analyses suggested that surgery conferred a survival advantage among women with solitary bone metastases but a survival disadvantage among women with multiple liver or lung metastases.

The rate of locoregional progression was one-fifth as high with surgery as with systemic therapy.

"There was no statistically significant difference in overall survival at early follow-up, but we need longer follow-up," Dr. Soran commented. "There were potentially important subgroup differences."

The two trials have implications – both for other ongoing trials and for patient care – according to invited discussant Dr. Seema A. Khan, a professor of surgery and Bluhm Family Professor of Cancer Research at Northwestern University, Chicago, and a physician at the university’s Lynn Sage Breast Center.

"A large benefit of primary site local therapy seems unlikely based on the data we saw today," she maintained. "The assumptions we have used in our ongoing trial designs will need to be reassessed. Preplanned pooled analyses may yield sufficient power to detect smaller differences."

As for patient care, "it is pretty clear that at this point in time, we have to make sure that our patients understand that there is really no proven survival advantage to primary site local therapy," Dr. Khan said. "So I don’t think this treatment should be offered to patients with asymptomatic tumors unless they are participating in a clinical trial. There may be a local control advantage; we need to see more data on that."

Indian trial

Participants in the Indian trial, conducted between 2005 and 2013, were women with de novo metastatic breast cancer who had had a complete or partial response to anthracycline chemotherapy alone or with a taxane.

They were randomized evenly to receive locoregional therapy (lumpectomy or mastectomy with axillary lymph node dissection, plus radiation therapy to the chest wall or breast and lymph nodes) or no locoregional therapy as a control. Both groups received hormonal therapy if indicated.

In the control group, about 10% of women underwent a palliative mastectomy because of impending fungation or pain in the breast, as permitted by study protocol, according to Dr. Badwe.

The patients thus received contemporary therapy, he said, except that the 16% with HER2-positive disease did not receive the targeted agent trastuzumab (Herceptin).

Results showed that median overall survival was about 18 months, and the 5-year rate was 19.2% with locoregional therapy and 20.5% without it, a nonsignificant difference. "Uniformly, there was no difference at all in any subsets," Dr. Badwe reported.

The study had a one-sided superiority design, he noted. "We wouldn’t have been able to see a 2.5- or 3-month difference, or a 4% detriment" in overall survival.

The surgery group had a lower risk of local progression-free survival events (hazard ratio, 0.16; P = .00), but a higher risk of distant progression-free survival events (HR, 1.42; P = .01).

Several theories might explain why removal of the primary would accelerate growth of metastases, according to Dr. Badwe.

"The first and the foremost is that the act of surgery itself might elaborate some growth factors which might allow metastatic disease to grow. The second possibility, which was suggested by Dr. Fisher (Cancer Res. 1989;49:1996-2001) is that the primary tumor, which predates the onset of distant metastases, elaborates some inhibitory factors, and they are not there once the primary tumor is removed, bestowing autonomy of growth on the distant metastases," he explained. "And the third possibility is the act of surgery might induce some more metastatic processes by dissemination and create new disease."

Session attendee Dr. Steven Vogl, an oncologist in the Bronx, N.Y., said, "The ascertainment of disease progression requires disease that you can follow. A woman with bone-only metastases with her cancer in place, you can tell when her cancer is getting worse because the primary site or the axillary nodes are getting bigger. It’s much more difficult if you’ve taken those off and irradiated the chest wall. Have you looked at your data to see if that’s what was going on, why you had more distant metastases, because they couldn’t progress locally? This is a medical trial explanation that contradicts Fisher’s biologic hypothesis."

"There was a fixed time duration at which systemic investigations were performed to assess whether the distant metastases progressed," Dr. Badwe replied. If anything, the patients who did not have surgery had more assessments of their distant metastases, he said.

Dr. Tari A. King, a session attendee from the Memorial Sloan-Kettering Cancer Center, New York, noted that a lack of HER2 therapy in the trial may have had a large effect.

"We do have prospective registry data here [abstract 18-09, presented in a poster session] from a trial that we completed in the United States sponsored by the Translational Breast Cancer Research Consortium, and the patients in our study, whether they received surgery or not, their 2-year overall survival is far superior to what you’ve just showed us," she commented. "So I’m not sure that we can really apply your data to the modern targeted therapy regimens that we see in the United States."

Turkish trial

The Turkish trial, known as the MF07-01 trial, was conduced between 2008 and 2012 among treatment-naïve patients.

They were randomized evenly to receive either systemic therapy alone or surgery for the primary tumor (with or without axillary dissection) followed by radiation therapy if indicated, plus systemic therapy.

All patients received hormonal therapy as needed, and those with HER2-positive disease received trastuzumab.

With a median follow-up of 18 months, the median overall survival was 46 months with initial surgery and 42 months with initial systemic therapy, a nonsignificant difference, reported Dr. Soran, who disclosed no relevant conflicts of interest.

In unplanned subgroup analyses, the findings were similar for most subgroups of patients. However, surgery yielded superior survival in patients with solitary bone metastases (not reached vs. 42 months, P = .02) and inferior survival in patients having multiple liver or pulmonary metastases (16 months vs. not reached, P = .02).

The rate of locoregional progression was much lower with initial surgery than with initial systemic therapy (0.7% vs. 3.6%).

Dr. Soran emphasized that the trial’s planned median follow-up is 36 months, so the presented results are only preliminary. Quality of life and morbidity analyses are ongoing.

Dr. Badwe and Dr. Soran disclosed no relevant conflicts of interest.

SAN ANTONIO – Surgical removal of the primary tumor and affected lymph nodes afforded no overall survival benefit in women who had metastatic breast cancer at initial presentation, according to a pair of randomized trials presented at the San Antonio Breast Cancer Symposium.

In a first trial, conducted among 350 women in India who had responded to initial chemotherapy, about 20% of the women were still alive at 5 years regardless of whether they had surgery or just received more systemic therapy, first author Dr. Rajendra Badwe, director of the Tata Memorial Hospital in Mumbai, India, reported in a session and at a press briefing.

Superior locoregional control conferred by surgery was canceled out by a higher risk of progression in distant sites, lending support to more than 20-year-old preclinical data by Dr. Bernard Fisher and his colleagues suggesting that an intact primary suppresses growth of distant metastases (Cancer Res. 1989;49:1996-2001).

"Locoregional treatment of the primary tumor in women presenting with metastatic breast cancer did not result in any overall survival benefit and hence should not be offered as a routine practice," Dr. Badwe commented.

"The biological fallout of this study is that surgical removal of the primary tumor in these women appears to confer a growth advantage on distant metastases. ... This is the first time that we have evidence in human studies that locoregional treatment has a great kinetic effect on distant metastases," he added.

In a second trial, conducted among 293 women in Turkey with untreated de novo metastatic breast cancer, median estimated survival was statistically indistinguishable, at about 3.5 years, regardless of whether women had up-front surgery or simply systemic therapy, first author Dr. Atilla Soran reported in a session on behalf of the Turkish Federation of Societies for Breast Diseases.

Subgroup analyses suggested that surgery conferred a survival advantage among women with solitary bone metastases but a survival disadvantage among women with multiple liver or lung metastases.

The rate of locoregional progression was one-fifth as high with surgery as with systemic therapy.

"There was no statistically significant difference in overall survival at early follow-up, but we need longer follow-up," Dr. Soran commented. "There were potentially important subgroup differences."

The two trials have implications – both for other ongoing trials and for patient care – according to invited discussant Dr. Seema A. Khan, a professor of surgery and Bluhm Family Professor of Cancer Research at Northwestern University, Chicago, and a physician at the university’s Lynn Sage Breast Center.

"A large benefit of primary site local therapy seems unlikely based on the data we saw today," she maintained. "The assumptions we have used in our ongoing trial designs will need to be reassessed. Preplanned pooled analyses may yield sufficient power to detect smaller differences."

As for patient care, "it is pretty clear that at this point in time, we have to make sure that our patients understand that there is really no proven survival advantage to primary site local therapy," Dr. Khan said. "So I don’t think this treatment should be offered to patients with asymptomatic tumors unless they are participating in a clinical trial. There may be a local control advantage; we need to see more data on that."

Indian trial

Participants in the Indian trial, conducted between 2005 and 2013, were women with de novo metastatic breast cancer who had had a complete or partial response to anthracycline chemotherapy alone or with a taxane.

They were randomized evenly to receive locoregional therapy (lumpectomy or mastectomy with axillary lymph node dissection, plus radiation therapy to the chest wall or breast and lymph nodes) or no locoregional therapy as a control. Both groups received hormonal therapy if indicated.

In the control group, about 10% of women underwent a palliative mastectomy because of impending fungation or pain in the breast, as permitted by study protocol, according to Dr. Badwe.

The patients thus received contemporary therapy, he said, except that the 16% with HER2-positive disease did not receive the targeted agent trastuzumab (Herceptin).

Results showed that median overall survival was about 18 months, and the 5-year rate was 19.2% with locoregional therapy and 20.5% without it, a nonsignificant difference. "Uniformly, there was no difference at all in any subsets," Dr. Badwe reported.

The study had a one-sided superiority design, he noted. "We wouldn’t have been able to see a 2.5- or 3-month difference, or a 4% detriment" in overall survival.

The surgery group had a lower risk of local progression-free survival events (hazard ratio, 0.16; P = .00), but a higher risk of distant progression-free survival events (HR, 1.42; P = .01).

Several theories might explain why removal of the primary would accelerate growth of metastases, according to Dr. Badwe.

"The first and the foremost is that the act of surgery itself might elaborate some growth factors which might allow metastatic disease to grow. The second possibility, which was suggested by Dr. Fisher (Cancer Res. 1989;49:1996-2001) is that the primary tumor, which predates the onset of distant metastases, elaborates some inhibitory factors, and they are not there once the primary tumor is removed, bestowing autonomy of growth on the distant metastases," he explained. "And the third possibility is the act of surgery might induce some more metastatic processes by dissemination and create new disease."

Session attendee Dr. Steven Vogl, an oncologist in the Bronx, N.Y., said, "The ascertainment of disease progression requires disease that you can follow. A woman with bone-only metastases with her cancer in place, you can tell when her cancer is getting worse because the primary site or the axillary nodes are getting bigger. It’s much more difficult if you’ve taken those off and irradiated the chest wall. Have you looked at your data to see if that’s what was going on, why you had more distant metastases, because they couldn’t progress locally? This is a medical trial explanation that contradicts Fisher’s biologic hypothesis."

"There was a fixed time duration at which systemic investigations were performed to assess whether the distant metastases progressed," Dr. Badwe replied. If anything, the patients who did not have surgery had more assessments of their distant metastases, he said.

Dr. Tari A. King, a session attendee from the Memorial Sloan-Kettering Cancer Center, New York, noted that a lack of HER2 therapy in the trial may have had a large effect.

"We do have prospective registry data here [abstract 18-09, presented in a poster session] from a trial that we completed in the United States sponsored by the Translational Breast Cancer Research Consortium, and the patients in our study, whether they received surgery or not, their 2-year overall survival is far superior to what you’ve just showed us," she commented. "So I’m not sure that we can really apply your data to the modern targeted therapy regimens that we see in the United States."

Turkish trial

The Turkish trial, known as the MF07-01 trial, was conduced between 2008 and 2012 among treatment-naïve patients.

They were randomized evenly to receive either systemic therapy alone or surgery for the primary tumor (with or without axillary dissection) followed by radiation therapy if indicated, plus systemic therapy.

All patients received hormonal therapy as needed, and those with HER2-positive disease received trastuzumab.

With a median follow-up of 18 months, the median overall survival was 46 months with initial surgery and 42 months with initial systemic therapy, a nonsignificant difference, reported Dr. Soran, who disclosed no relevant conflicts of interest.

In unplanned subgroup analyses, the findings were similar for most subgroups of patients. However, surgery yielded superior survival in patients with solitary bone metastases (not reached vs. 42 months, P = .02) and inferior survival in patients having multiple liver or pulmonary metastases (16 months vs. not reached, P = .02).

The rate of locoregional progression was much lower with initial surgery than with initial systemic therapy (0.7% vs. 3.6%).

Dr. Soran emphasized that the trial’s planned median follow-up is 36 months, so the presented results are only preliminary. Quality of life and morbidity analyses are ongoing.

Dr. Badwe and Dr. Soran disclosed no relevant conflicts of interest.