Metastatic breast cancer is generally considered incurable, and this colors doctor-patient interactions for patients with metastatic disease. Although true for most patients, there appear to be important exceptions, instances where long-term disease-free survival occurs. Although these instances are few in number, they suggest the possibility of cure. How will we move toward cure for a much larger population of patients with metastatic disease? This article outlines a potential research agenda that might move us toward that distant goal.
It may seem the height of presumption to speak of a cure for metastatic breast cancer, a disease stubbornly resistant to the best efforts of modern medicine. Oncologists are regularly cautioned to counsel patients with metastatic disease that their cancer is incurable, that all therapy is palliative in nature, and that there is little prospect for long-term survival. Statistics seem to bear this out: the American Cancer Society reports 5-year survival rates for metastatic breast cancer of 22%, not a number of which to be proud.1 Similarly, although many articles have suggested gradually improving survival for patients with metastatic breast cancer, a recent analysis by the Eastern Cooperative Oncology Group of its large phase III trials has suggested that, adjusted for distant relapse-free survival, metastatic survival has barely budged in recent decades.2
All of these statements are true for the great majority of patients, but not, even today, for all. Nor should we assume that past performance predicts future returns. In fact, there is reasonable evidence to suggest that some patients with metastatic breast cancer are cured, albeit currently a small number.3 And if some patients are cured, might not we cure more?
WHAT DO WE MEAN BY CURE?
Cure would seem to be a simple thing to define, but in actuality it has several possible definitions. A statistical cure can be said to occur when the progression-free survival curve for a metastatic population undergoes a plateau after therapy or, alternatively, when the overall survival for a metastatic population approaches that of an age- and sex-matched general population; a personal cure refers to an individual patient who, while in apparent remission, dies of some disease other than breast cancer. Both are of value, but differ significantly: a patient with a complete response to metastatic chemotherapy who dies in an automobile accident might, of course, have relapsed in short order.
Similarly, does cure include conversion of metastatic breast cancer to a chronic disease? The cancer as a chronic disease meme has been an important one in recent years, embraced by both physicians and patients. It draws from sources such as diabetes mellitus, where a disease is held in check, at little cost in the way of disease or treatment-related symptoms, for decades. As will be discussed later in this review, the advent of novel immunotherapy approaches such as checkpoint inhibition suggests that cancer as a chronic disease may represent a real possibility. However, the usual path of metastatic breast cancer, wherein patients may receive sequential drugs over years, even decades, hardly seems to fall under what a reasonable person would think of as a chronic disease. Current metastatic therapies are expensive, often toxic, and doomed to the eventual development of drug resistance.
One argument against therapeutic nihilism is that a small but real fraction of patients with metastatic disease are, in fact, currently curable.3 The evidence supporting this suggestion derives from multiple sources (Table 1).
Table
Table 1. Lines of Evidence Suggesting Metastatic Breast Cancer Is Curable
Table 1. Lines of Evidence Suggesting Metastatic Breast Cancer Is Curable
Adjuvant therapy cures micrometastasis
Adjuvant therapy after isolated local-regional recurrence improves survival
Chemotherapy for overt metastatic disease produces long-term survivors
Exceptional responders with novel agents
Treatment of low-volume metastatic disease with surgery and radiation produces long-term survivors
Adjuvant Therapy
Adjuvant therapy, by definition, treats metastatic disease. That the metastatic disease treated is microscopic does not in any way refute the notion that a long-term survivor has had metastatic disease cured with systemic therapy. That the same patient, treated with the same regimen for overt metastatic disease, is incurable suggests only that there is some boundary limit (based on either size or biology) beyond which that patient’s cancer becomes incurable.
Adjuvant Therapy After Isolated Local-Regional Recurrence
A generation of oncologists understood that local-regional recurrences were harbingers of overt metastatic disease, which frequently followed them in short order. The recent results of the CALOR (Chemotherapy as Adjuvant for Locally Recurrent Breast Cancer) trial, which demonstrate that chemotherapy given after local-regional therapy is associated with a reduced metastatic death rate, suggest that at least some patients with metastatic disease are likely cured by such pseudo-adjuvant therapy.4 Again, as with adjuvant therapy, tumor burden and biology (estrogen receptor–negative patients appear to derive greater benefit than estrogen receptor–positive patients) are important predictors of benefit.
Metastatic Chemotherapy
Data from trials of anthracycline-based chemotherapy for metastatic breast cancer, conducted at MD Anderson Cancer Center, suggest that a small but real number of patients—perhaps in the range of 1% to 2%—are long-term (> 10 years) disease-free survivors.5 Similarly, in the analysis of Eastern Cooperative Oncology Group phase II trials mentioned above, a small but real fraction of long-term (> 15 years) survivors occurs across essentially all trials.2 Although many such patients might have been cured with modern adjuvant chemotherapy, the existence of such long-term survivors suggests that some patients are, indeed, cured of their metastatic disease with combination chemotherapy. In general, long-term survivors tend to combine good performance status, younger age, and relatively low volumes of disease.
Exceptional Responders
A common finding of many metastatic studies, with either standard or investigational therapies, is the occurrence of rare long-term survivors. These events are seen with both standard and investigational therapies. Their rare occurrence in the setting of investigational agents often dooms such drugs to the trash pile of oncologic history. Presumably something ultimately identifiable in their tumor biology will explain therapeutic benefit.
Treatment of Low-Volume Metastatic Disease
Numerous studies (reviewed by Kobayashi et al6 and Cheng and Ueno7) have suggested that some patients with low-volume (primarily oligometastatic) disease may be treated with curative intent. Emblematic of these is the study by Kobayashi et al6 of 75 patients with oligometastatic disease (defined as one or two organs involved with metastatic lesions [excluding the primary lesion resectable by surgery], fewer than five lesions per metastasized organ, and lesion diameter < 5 cm). Patients were treated with systemic therapy to maximum response, then with local therapy (radiation or surgery) where appropriate. In their analysis, patients with single organ involvement had a 52% overall survival at 20 years. Long-term survivors, unsurprisingly, were those with a complete response to combined modality therapy.
The existence of such long-term survivors does not represent a prescription for combined modality therapy in all patients with metastatic disease. This literature is confounded by the lack of prospectively agreed on objective criteria for the use of local therapies (surgery and radiotherapy) and, in many studies, the lack of long-term follow-up. What percentage of patients receiving such an approach are long-term survivors? We simply do not know. But there are clearly some.
What all of these cases seem to share in common is the presence of low-volume disease; quantity has a quality all its own. Size kills. This may represent biology as much as simple volume: larger tumors, having undergone more cell divisions, have had both greater opportunity to develop more lethal mutations (assuming a constant mutation rate per cell division) and more time to generate secondary sites of metastasis.
As the great 20th century philosopher Yogi Berra once opined, prediction is difficult, especially about the future. Are there any likely paths to an improved cure rate? Several prospects suggest themselves and collectively represent a potential research agenda for the cure of metastatic breast cancer (Table 2). Based on what we know of the biology of breast cancer, it would be unlikely that any individual approach would be successful in more than a fraction of patients.
Table
Table 2. A Research Agenda for Curing Metastatic Disease
Table 2. A Research Agenda for Curing Metastatic Disease
Earlier detection of metastatic breast cancer through better imaging
Early detection of treatable genomic lesions with ctDNA
Generate numerous exceptional responders
Target genomically smart cancers with immunotherapeutic approaches
Abbreviation: ctDNA, circulating tumor DNA.
Earlier Detection of Metastatic Breast Cancer Through Better Imaging
If, as suggested above, there comes a point in the natural history of metastatic breast cancer (for which size is a ready surrogate) when a cancer goes from being potentially curable to regularly incurable, then one potential point of attack would seem to be the earlier detection of metastatic disease. Earlier detection might come about as a result of improved imaging resolution.
It might well be argued that we have been there, done that in the 1990s, when two well-conducted randomized surveillance trials of patients with early-stage disease failed to demonstrate a survival benefit for early detection of disease recurrence.8,9 No new randomized controlled trials evaluating a survival end point have been performed in the surveillance space since the 1990s. Indeed, ASCO has not measurably changed its surveillance recommendations for patients with early-stage disease since 1997.10
In reality, however, the surveillance technologies available at the time (eg, bone scans, liver ultrasounds, chest x-rays) were crude by today’s standards and likely totally incomparable to the techniques that will become available over the next decade. For instance, positron emission tomography/computed tomography scans have not been examined in a randomized controlled trial for their effect on outcomes in the surveillance setting, although a multitude of small, retrospective, uncontrolled trials have been performed.11 The CALOR trial, which used local-regional recurrence as a surrogate for the existence of metastatic disease, clearly suggests that late adjuvant/early metastatic chemotherapy provides potentially curative benefit for some such patients.4
Imaging resolution continues to improve and may be in the submillimeter range within the next decade; experimental approaches have already reached this size range.12,13 Of course, finding submillimeter lung or liver metastases may represent a mixed blessing: how will we know which ones are likely to progress to overt metastatic disease? And if we do not know how to cure such lesions, will we have accomplished anything other than increasing patient anxiety and health care costs? Nevertheless, the development of submillimeter imaging for recurrent disease offers the possibility of developing trials aimed at curing early metastatic cancer.14
Early Detection of Treatable Genomic Lesions
One reason size kills, as is suggested above, is that it is associated with increased tumor heterogeneity: tumors simply become smarter over time as they accumulate mutational events. Early detection, not just of micrometastatic disease but of early mutational events leading to drug resistance, might allow introduction of novel agents targeting those resistance mechanisms. Measuring circulating tumor DNA (ctDNA) for such emerging lesions is already possible, although no therapeutic trials based on such observations have yet been initiated in breast cancer.15
Simply detecting tumor-specific ctDNA is unlikely to suffice; the lead time between ctDNA detection and overt metastatic disease is, at present, relatively short (8 months in one recent study of patients with breast cancer). This is little different from the lead time seen with the use of protein biomarkers such as CA27-29.16 Finding bad news at an earlier point in time does not automatically turn it into good news, as the Southwest Oncology Group’s S007 circulating tumor trial demonstrated.17 What might be crucial about ctDNA is not lead time but rather its potential for determining effective therapy for a (micro/early) metastatic lesion.
Generating Large Numbers of Exceptional Responders
Exceptional responders, as defined by the National Cancer Institute (NCI), are by definition rare; in the terms used by the NCI these are patients who received a treatment in which fewer than 10% of patients had a complete response or a durable (lasting at least 6 months) partial response based on clinical trial data or extensive historical experience in the context of the patient’s tumor type, or achieved either a complete response or a partial response with duration of at least 6 months as defined by RECIST (Response Evaluation Criteria in Solid Tumors) criteria.
Would a systematic analysis of these exceptional responders generate a significant cumulative percentage of patients deriving great benefit from a large palette of drugs, many never considered for patients with advanced disease? We have no answer to this question at present, although the NCI’s Exceptional Responder Study (NCT02243592) is a first attempt to provide an answer.
DNA and RNA from tissue samples taken from exceptional responders will be isolated and then undergo whole-exome sequencing and/or mRNA sequencing. The hope is that analysis of exceptional responders “could help identify novel or known molecular abnormalities that are likely to be predictive of response to agents with targets in a relevant biological pathway.”18
Numerous ongoing basket trials are currently attempting to use genomic analyses to inform treatment decisions. To date, few of these trials have analyzed more than a handful of patients with metastatic breast cancer. Similarly, none have suggested that detection of a single molecular lesion in the metastatic setting will render a patient curable. One recent randomized phase II trial (albeit in a general metastatic cancer population) has suggested that a genomically driven treatment is no better than a doctor’s best choice regimen for the average patient.19
Indeed, it is unlikely that such would be the case: a genomic analysis of 100 primary breast cancers suggested that single driver mutations were found in only 28% of cases, and some cancers had as many as six driver mutations.20 Breast cancer metastases are more genetically diverse than primary tumors.21 Similarly, recent data looking at single-cell sequencing in patients with breast cancer reveals an astonishing degree of molecular heterogeneity. In the frightening words of the article, “no two cancer cells are identical.”22(p158) So, although worth pursuing, a genomic attack on metastatic cancer is unlikely to represent a panacea, though incremental gains may occur.
One important counterargument to both the early detection of treatable genomic lesions and the generation of large numbers of exceptional responders is that many metastatic breast cancers are genomically smart tumors (ie, they have accumulated large numbers of mutations during their natural history). Indeed, many breast cancers are highly genomically heterogenous at an early stage.22 Their genomic diversity, in essence, predestines them to multidrug resistance.
Immunotherapeutic Approaches
For such genomically smart cancers, the rapidly emerging field of cancer immunotherapy represents a possible approach to increasing long-term survival. Analyses of checkpoint inhibitor therapies have suggested that the most heavily mutagenized cancers (ie, the ones least likely to respond to standard kinase-based targeted therapies) are paradoxically the ones most likely to respond to checkpoint inhibition with anti–CTLA-4 or anti–PD-1/PD-L1 antibody therapy.23 Increased mutagenesis is associated with increased numbers of immunogenic neoantigens.
Immunotherapeutic approaches involving checkpoint inhibitor therapy are in their infancy in breast cancer, with only a few dozen patients reported to date.24,25 These trials have, by and large, involved patients with so-called triple-negative breast cancers. These patients represent a rational starting point for checkpoint inhibitors: they are both the most highly mutagenized cancers and the cancers most likely to have demonstrable immune T-cell infiltrates (tumor-infiltrating lymphocytes). Studies of tumor-infiltrating lymphocytes in the adjuvant setting have suggested that their presence is a powerful predictor of outcome, suggesting that we already rely on the immune system for therapeutic impact even in the absence of immune-based therapies.26
But immunotherapeutic approaches to breast cancer, indeed all cancers, are clearly in their infancy. Other novel approaches abound, some of which may extend immunotherapy’s reach beyond triple-negative disease. Where we will be a decade from now may be almost unimaginably different. Whether such patients will have residual measurable metastatic disease (eg, a liver metastasis held in check by the body’s reactivated immune system), and if so whether such a patient can truly be said to be cured, are interesting questions. Interesting, but of uncertain importance: if such a patient has a personal cure, dying 20 years later of some noncancer disease, she may not care that her computed tomography scan is subtly marred, if she pays a minimal price with regard to toxicity and if treatment duration is brief.
In conclusion, it may appear Pollyanna-like to even consider the possibility of curing metastatic breast cancer. As I have argued, however, there are several potential paths to a cure; indeed, given the different diseases that make up what we call breast cancer, many paths will need to be trod on this long journey. We should avoid false promises (which have occurred all too frequently in our field), but we should also avoid nihilism. The journey is worth taking, even if the goal currently seems elusive.
AUTHOR’S DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Curing Metastatic Breast Cancer
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or jop.ascopubs.org/site/misc/ifc.xhtml.
It may seem the height of presumption to speak of a cure for metastatic breast cancer, a disease stubbornly resistant to the best efforts of modern medicine. Oncologists are regularly cautioned to counsel patients with metastatic disease that their cancer is incurable, that all therapy is palliative in nature, and that there is little prospect for long-term survival. Statistics seem to bear this out: the American Cancer Society reports 5-year survival rates for metastatic breast cancer of 22%, not a number of which to be proud.1 Similarly, although many articles have suggested gradually improving survival for patients with metastatic breast cancer, a recent analysis by the Eastern Cooperative Oncology Group of its large phase III trials has suggested that, adjusted for distant relapse-free survival, metastatic survival has barely budged in recent decades.2
All of these statements are true for the great majority of patients, but not, even today, for all. Nor should we assume that past performance predicts future returns. In fact, there is reasonable evidence to suggest that some patients with metastatic breast cancer are cured, albeit currently a small number.3 And if some patients are cured, might not we cure more?
WHAT DO WE MEAN BY CURE?
Cure would seem to be a simple thing to define, but in actuality it has several possible definitions. A statistical cure can be said to occur when the progression-free survival curve for a metastatic population undergoes a plateau after therapy or, alternatively, when the overall survival for a metastatic population approaches that of an age- and sex-matched general population; a personal cure refers to an individual patient who, while in apparent remission, dies of some disease other than breast cancer. Both are of value, but differ significantly: a patient with a complete response to metastatic chemotherapy who dies in an automobile accident might, of course, have relapsed in short order.
Similarly, does cure include conversion of metastatic breast cancer to a chronic disease? The cancer as a chronic disease meme has been an important one in recent years, embraced by both physicians and patients. It draws from sources such as diabetes mellitus, where a disease is held in check, at little cost in the way of disease or treatment-related symptoms, for decades. As will be discussed later in this review, the advent of novel immunotherapy approaches such as checkpoint inhibition suggests that cancer as a chronic disease may represent a real possibility. However, the usual path of metastatic breast cancer, wherein patients may receive sequential drugs over years, even decades, hardly seems to fall under what a reasonable person would think of as a chronic disease. Current metastatic therapies are expensive, often toxic, and doomed to the eventual development of drug resistance.
THE PRESENT-DAY CURABILITY OF METASTATIC BREAST CANCER
One argument against therapeutic nihilism is that a small but real fraction of patients with metastatic disease are, in fact, currently curable.3 The evidence supporting this suggestion derives from multiple sources (Table 1).
Table
Table 1. Lines of Evidence Suggesting Metastatic Breast Cancer Is Curable
Table 1. Lines of Evidence Suggesting Metastatic Breast Cancer Is Curable
Adjuvant therapy cures micrometastasis
Adjuvant therapy after isolated local-regional recurrence improves survival
Chemotherapy for overt metastatic disease produces long-term survivors
Exceptional responders with novel agents
Treatment of low-volume metastatic disease with surgery and radiation produces long-term survivors
Adjuvant Therapy
Adjuvant therapy, by definition, treats metastatic disease. That the metastatic disease treated is microscopic does not in any way refute the notion that a long-term survivor has had metastatic disease cured with systemic therapy. That the same patient, treated with the same regimen for overt metastatic disease, is incurable suggests only that there is some boundary limit (based on either size or biology) beyond which that patient’s cancer becomes incurable.
Adjuvant Therapy After Isolated Local-Regional Recurrence
A generation of oncologists understood that local-regional recurrences were harbingers of overt metastatic disease, which frequently followed them in short order. The recent results of the CALOR (Chemotherapy as Adjuvant for Locally Recurrent Breast Cancer) trial, which demonstrate that chemotherapy given after local-regional therapy is associated with a reduced metastatic death rate, suggest that at least some patients with metastatic disease are likely cured by such pseudo-adjuvant therapy.4 Again, as with adjuvant therapy, tumor burden and biology (estrogen receptor–negative patients appear to derive greater benefit than estrogen receptor–positive patients) are important predictors of benefit.
Metastatic Chemotherapy
Data from trials of anthracycline-based chemotherapy for metastatic breast cancer, conducted at MD Anderson Cancer Center, suggest that a small but real number of patients—perhaps in the range of 1% to 2%—are long-term (> 10 years) disease-free survivors.5 Similarly, in the analysis of Eastern Cooperative Oncology Group phase II trials mentioned above, a small but real fraction of long-term (> 15 years) survivors occurs across essentially all trials.2 Although many such patients might have been cured with modern adjuvant chemotherapy, the existence of such long-term survivors suggests that some patients are, indeed, cured of their metastatic disease with combination chemotherapy. In general, long-term survivors tend to combine good performance status, younger age, and relatively low volumes of disease.
Exceptional Responders
A common finding of many metastatic studies, with either standard or investigational therapies, is the occurrence of rare long-term survivors. These events are seen with both standard and investigational therapies. Their rare occurrence in the setting of investigational agents often dooms such drugs to the trash pile of oncologic history. Presumably something ultimately identifiable in their tumor biology will explain therapeutic benefit.
Treatment of Low-Volume Metastatic Disease
Numerous studies (reviewed by Kobayashi et al6 and Cheng and Ueno7) have suggested that some patients with low-volume (primarily oligometastatic) disease may be treated with curative intent. Emblematic of these is the study by Kobayashi et al6 of 75 patients with oligometastatic disease (defined as one or two organs involved with metastatic lesions [excluding the primary lesion resectable by surgery], fewer than five lesions per metastasized organ, and lesion diameter < 5 cm). Patients were treated with systemic therapy to maximum response, then with local therapy (radiation or surgery) where appropriate. In their analysis, patients with single organ involvement had a 52% overall survival at 20 years. Long-term survivors, unsurprisingly, were those with a complete response to combined modality therapy.
The existence of such long-term survivors does not represent a prescription for combined modality therapy in all patients with metastatic disease. This literature is confounded by the lack of prospectively agreed on objective criteria for the use of local therapies (surgery and radiotherapy) and, in many studies, the lack of long-term follow-up. What percentage of patients receiving such an approach are long-term survivors? We simply do not know. But there are clearly some.
What all of these cases seem to share in common is the presence of low-volume disease; quantity has a quality all its own. Size kills. This may represent biology as much as simple volume: larger tumors, having undergone more cell divisions, have had both greater opportunity to develop more lethal mutations (assuming a constant mutation rate per cell division) and more time to generate secondary sites of metastasis.
THE PROSPECTS FOR INCREASING THE CURE RATE FOR METASTATIC BREAST CANCER: AN AGENDA
As the great 20th century philosopher Yogi Berra once opined, prediction is difficult, especially about the future. Are there any likely paths to an improved cure rate? Several prospects suggest themselves and collectively represent a potential research agenda for the cure of metastatic breast cancer (Table 2). Based on what we know of the biology of breast cancer, it would be unlikely that any individual approach would be successful in more than a fraction of patients.
Table
Table 2. A Research Agenda for Curing Metastatic Disease
Table 2. A Research Agenda for Curing Metastatic Disease
Earlier detection of metastatic breast cancer through better imaging
Early detection of treatable genomic lesions with ctDNA
Generate numerous exceptional responders
Target genomically smart cancers with immunotherapeutic approaches
Abbreviation: ctDNA, circulating tumor DNA.
Earlier Detection of Metastatic Breast Cancer Through Better Imaging
If, as suggested above, there comes a point in the natural history of metastatic breast cancer (for which size is a ready surrogate) when a cancer goes from being potentially curable to regularly incurable, then one potential point of attack would seem to be the earlier detection of metastatic disease. Earlier detection might come about as a result of improved imaging resolution.
It might well be argued that we have been there, done that in the 1990s, when two well-conducted randomized surveillance trials of patients with early-stage disease failed to demonstrate a survival benefit for early detection of disease recurrence.8,9 No new randomized controlled trials evaluating a survival end point have been performed in the surveillance space since the 1990s. Indeed, ASCO has not measurably changed its surveillance recommendations for patients with early-stage disease since 1997.10
In reality, however, the surveillance technologies available at the time (eg, bone scans, liver ultrasounds, chest x-rays) were crude by today’s standards and likely totally incomparable to the techniques that will become available over the next decade. For instance, positron emission tomography/computed tomography scans have not been examined in a randomized controlled trial for their effect on outcomes in the surveillance setting, although a multitude of small, retrospective, uncontrolled trials have been performed.11 The CALOR trial, which used local-regional recurrence as a surrogate for the existence of metastatic disease, clearly suggests that late adjuvant/early metastatic chemotherapy provides potentially curative benefit for some such patients.4
Imaging resolution continues to improve and may be in the submillimeter range within the next decade; experimental approaches have already reached this size range.12,13 Of course, finding submillimeter lung or liver metastases may represent a mixed blessing: how will we know which ones are likely to progress to overt metastatic disease? And if we do not know how to cure such lesions, will we have accomplished anything other than increasing patient anxiety and health care costs? Nevertheless, the development of submillimeter imaging for recurrent disease offers the possibility of developing trials aimed at curing early metastatic cancer.14
Early Detection of Treatable Genomic Lesions
One reason size kills, as is suggested above, is that it is associated with increased tumor heterogeneity: tumors simply become smarter over time as they accumulate mutational events. Early detection, not just of micrometastatic disease but of early mutational events leading to drug resistance, might allow introduction of novel agents targeting those resistance mechanisms. Measuring circulating tumor DNA (ctDNA) for such emerging lesions is already possible, although no therapeutic trials based on such observations have yet been initiated in breast cancer.15
Simply detecting tumor-specific ctDNA is unlikely to suffice; the lead time between ctDNA detection and overt metastatic disease is, at present, relatively short (8 months in one recent study of patients with breast cancer). This is little different from the lead time seen with the use of protein biomarkers such as CA27-29.16 Finding bad news at an earlier point in time does not automatically turn it into good news, as the Southwest Oncology Group’s S007 circulating tumor trial demonstrated.17 What might be crucial about ctDNA is not lead time but rather its potential for determining effective therapy for a (micro/early) metastatic lesion.
Generating Large Numbers of Exceptional Responders
Exceptional responders, as defined by the National Cancer Institute (NCI), are by definition rare; in the terms used by the NCI these are patients who received a treatment in which fewer than 10% of patients had a complete response or a durable (lasting at least 6 months) partial response based on clinical trial data or extensive historical experience in the context of the patient’s tumor type, or achieved either a complete response or a partial response with duration of at least 6 months as defined by RECIST (Response Evaluation Criteria in Solid Tumors) criteria.
Would a systematic analysis of these exceptional responders generate a significant cumulative percentage of patients deriving great benefit from a large palette of drugs, many never considered for patients with advanced disease? We have no answer to this question at present, although the NCI’s Exceptional Responder Study (NCT02243592) is a first attempt to provide an answer.
DNA and RNA from tissue samples taken from exceptional responders will be isolated and then undergo whole-exome sequencing and/or mRNA sequencing. The hope is that analysis of exceptional responders “could help identify novel or known molecular abnormalities that are likely to be predictive of response to agents with targets in a relevant biological pathway.”18
Numerous ongoing basket trials are currently attempting to use genomic analyses to inform treatment decisions. To date, few of these trials have analyzed more than a handful of patients with metastatic breast cancer. Similarly, none have suggested that detection of a single molecular lesion in the metastatic setting will render a patient curable. One recent randomized phase II trial (albeit in a general metastatic cancer population) has suggested that a genomically driven treatment is no better than a doctor’s best choice regimen for the average patient.19
Indeed, it is unlikely that such would be the case: a genomic analysis of 100 primary breast cancers suggested that single driver mutations were found in only 28% of cases, and some cancers had as many as six driver mutations.20 Breast cancer metastases are more genetically diverse than primary tumors.21 Similarly, recent data looking at single-cell sequencing in patients with breast cancer reveals an astonishing degree of molecular heterogeneity. In the frightening words of the article, “no two cancer cells are identical.”22(p158) So, although worth pursuing, a genomic attack on metastatic cancer is unlikely to represent a panacea, though incremental gains may occur.
One important counterargument to both the early detection of treatable genomic lesions and the generation of large numbers of exceptional responders is that many metastatic breast cancers are genomically smart tumors (ie, they have accumulated large numbers of mutations during their natural history). Indeed, many breast cancers are highly genomically heterogenous at an early stage.22 Their genomic diversity, in essence, predestines them to multidrug resistance.
Immunotherapeutic Approaches
For such genomically smart cancers, the rapidly emerging field of cancer immunotherapy represents a possible approach to increasing long-term survival. Analyses of checkpoint inhibitor therapies have suggested that the most heavily mutagenized cancers (ie, the ones least likely to respond to standard kinase-based targeted therapies) are paradoxically the ones most likely to respond to checkpoint inhibition with anti–CTLA-4 or anti–PD-1/PD-L1 antibody therapy.23 Increased mutagenesis is associated with increased numbers of immunogenic neoantigens.
Immunotherapeutic approaches involving checkpoint inhibitor therapy are in their infancy in breast cancer, with only a few dozen patients reported to date.24,25 These trials have, by and large, involved patients with so-called triple-negative breast cancers. These patients represent a rational starting point for checkpoint inhibitors: they are both the most highly mutagenized cancers and the cancers most likely to have demonstrable immune T-cell infiltrates (tumor-infiltrating lymphocytes). Studies of tumor-infiltrating lymphocytes in the adjuvant setting have suggested that their presence is a powerful predictor of outcome, suggesting that we already rely on the immune system for therapeutic impact even in the absence of immune-based therapies.26
But immunotherapeutic approaches to breast cancer, indeed all cancers, are clearly in their infancy. Other novel approaches abound, some of which may extend immunotherapy’s reach beyond triple-negative disease. Where we will be a decade from now may be almost unimaginably different. Whether such patients will have residual measurable metastatic disease (eg, a liver metastasis held in check by the body’s reactivated immune system), and if so whether such a patient can truly be said to be cured, are interesting questions. Interesting, but of uncertain importance: if such a patient has a personal cure, dying 20 years later of some noncancer disease, she may not care that her computed tomography scan is subtly marred, if she pays a minimal price with regard to toxicity and if treatment duration is brief.
In conclusion, it may appear Pollyanna-like to even consider the possibility of curing metastatic breast cancer. As I have argued, however, there are several potential paths to a cure; indeed, given the different diseases that make up what we call breast cancer, many paths will need to be trod on this long journey. We should avoid false promises (which have occurred all too frequently in our field), but we should also avoid nihilism. The journey is worth taking, even if the goal currently seems elusive.
AUTHOR’S DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Curing Metastatic Breast Cancer
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or jop.ascopubs.org/site/misc/ifc.xhtml.
