Recommendations for streamlining precision medicine in breast cancer care in Latin America

Abstract Background The incidence of breast cancer (BC) in LMICs has increased by more than 20% within the last decade. In areas such as Latin America (LA), addressing BC at national levels evoke discussions surrounding fragmented care, limited resources, and regulatory barriers. Precision Medicine (PM), specifically companion diagnostics (CDx), links disease diagnosis and treatment for better patient outcomes. Thus, its application may aid in overcoming these barriers. Recent findings A panel of LA experts in fields related to BC and PM were provided with a series of relevant questions to address prior to a multi‐day conference. Within this conference, each narrative was edited by the entire group, through numerous rounds of discussion until a consensus was achieved. The panel proposes specific, realistic recommendations for implementing CDx in BC in LA and other LMIC regions. In these recommendations, the authors strived to address all barriers to the widespread use and access mentioned previously within this manuscript. Conclusion This manuscript provides a review of the current state of CDx for BC in LA. Of most importance, the panel proposes practical and actionable recommendations for the implementation of CDx throughout the Region in order to identify the right patient at the right time for the right treatment.

increase of BC incidence are changes in reproductive patterns (parity, breastfeeding), sedentary behavior, and unhealthy diet. 7 Young women comprise 30% of LA's population, which may explain the higher prevalence of triple-negative BC compared with other regions of the world, creating a larger burden of aggressive tumors. 8 HICs have more disability-adjusted life years (DALYs) for BC since they have higher general incidence rates, but the proportion of losses due to mortality are higher in LA. The mean number of DALYs lost due to BC in LA varies, being highest in Brazil, Mexico, and Argentina. 9 In order to establish a number of recommendations in relation to precision medicine (PM) in BC care in Latin America a panel of experts in this field were provided with a series of relevant questions to address prior to a multi-day conference. Within this conference, each narrative was edited by the entire group, through numerous rounds of discussion until a consensus was achieved.

| MULTIDISCIPLINARY BC CARE
Comprehensive cancer control involves prevention, early detection, diagnosis, treatment, rehabilitation, and palliative care. 10 In LA, most cancer care is carried out in a fragmented and sequential manner, 11,12 which contrasts with the contemporary standard of cancer care with a multidisciplinary team approach. 13 The lack of coordination in these models is aggravated by inadequate distribution of resources. Specialized units are disproportionately located in main cities, leaving large geographic areas underserved and limiting access to innovative therapies and technologies. Furthermore, educational efforts developed for medical personnel, especially in primary healthcare, can improve diagnosis, treatment, outcomes, and increase appropriate referrals. Patient organization involvement to support patients in ensuring compliance to follow-up care from the time of diagnosis may favor treatment continuity.
Data from LMICs, although limited, have shown the effectiveness of multidisciplinary cancer care. Evidence supports that tumor boards can aid in overcoming diagnostic and management barriers in limitedresource settings, where specialists may be less available. An alternative to this includes partnering with regional or private academic centers to carry out remote tumor boards, allowing for increased access to multidisciplinary expertise. 14 As molecular oncology increasingly becomes a part of standard of care, it is expected that molecular tumor boards will become as essential as disease-specific tumor boards are today. 15 A solution for achieving multidisciplinary care in BC is the implementation of Breast Units, which have been shown to improve overall patient outcomes. 16 In Breast Units, the multidisciplinary management approach involves radiologists, pathologists, radiation, surgical, and medical oncologists, nurses, plastic surgeons, psychologists, geneticists, and primary care physicians, among others. 17

| BC PATIENT JOURNEY
In order to visualize the usual flow of relevant processes a patient undergoes when diagnosed and treated for BC in LA, a conventional process mapping of the typical patient journey is illustrated in While the multidisciplinary approach is ideal, it is not available in most situations. In these cases, the patient usually starts the journey with a first consultation to a primary care doctor, followed by a surgical consultation for the evaluation of the suspicious breast condition.
After the histopathologic confirmation of malignancy, a treatment plan is defined by the clinical oncologist or surgeon who then refers the patient to other professionals as needed.
In this fragmented route, 18 coordination problems are frequent and delays in diagnosis and treatment are common. These may be due to several problems such as the absence of a structured process where patients receive support in their transit through the healthcare system and undergo proper monitoring, as well as the lack of education programs for first contact physician and patients. Frequent difficulties also exist regarding the quality of radiology and pathology reports and adherence to optimal standards of BC care. The patient journey is further complicated by reimbursement issues that obstruct access to medical care. 19 Clinical guidelines are necessary to streamline care and could result in cost and time savings, optimal use of resources, and improvement in the quality of patient care. They are available in many countries in LA, but they are not mandatory.

| IMPORTANCE OF EARLY DETECTION
Unequivocally, patients can live healthier and longer lives when provided a timely diagnosis and appropriate clinical management. 20,21 Early detection improves patient outcomes by increasing treatment options, increasing survival, and improving quality of life. The resulting diagnostic information empowers patients to better understand their prognosis and options on preventive action. Likewise, identifying genetic conditions or predispositions can be important for disease prevention and early management in patient relatives.
Cancer care varies within the different LA health systems. Financial constraints and administrative political decisions impinge cancer control, efficiency, and inclusiveness of overall cancer care. Although research on care delays and patient barriers have been conducted in select LA countries, such information has not been collated and synthesized in a standardized framework. Some reviews show common delays during every step of the cancer continuum, across cancer types, and country income-levels. 22,23 LA women face several access barriers to standard BC care when compared to those from developed countries. 12 This disparity is reflected in the considerable differences between the mortality and incidence ratios and the 5-year survival rates of LMICs and HICs. 10,24 A patient with advanced stage disease poses healthcare related costs up to three to four times more than a patient with stage I disease. 10 Evidence shows that early detection can produce "down staging" of the disease to stages that are more amenable to curative treatment and preventing early mortality. 21,22 Of note, while BC screening is an important prevention strategy, all levels of prevention should be addressed when examining solutions for better cancer care.
Mammography has remained the gold standard of BC screening throughout the world. To achieve optimal benefit, mammographic screening programs must be of high quality, appropriate targeting, and sufficient frequency. 25  Guidelines for BC screening in Mexico recommend mammographic screening every 2 years for women starting at 40 years of age, a practice that many of the Region's medical societies recommend. 7 Considering the high burden of aggressive tumors in young F I G U R E 1 Patient journey in Latin America patients, healthcare systems in LA may benefit from lowering age intervals of screening programs to 40 years of age. Evidence suggests that maximum cost-benefit can only occur if screening is done in an age group, which has a sufficiently high incidence of BC and high longevity. 26 Comprehensive research on compliance and effectivity of screening programs for the region and the cost-effectiveness of this practice are needed to support this notion.

| RESEARCH AND CANCER REGISTRIES
The main objective of cancer registries is to collect, code, and classify cancers to produce relevant statistics and provide a framework for assessing and controlling the burden of disease. In the majority of LA, there are no regional or countrywide registries for BC and the few that exist are exclusively in the initial stages of development. This is likely a result of social, cultural, and economic barriers between and within countries that make data collection difficult and costly. 27 An exception to this is Uruguay, which has The National Cancer Registry, a class A registry, created by law since 1989. 28,29 Establishing population-based cancer registries (PBCRs) should be a fundamental part of any national cancer control plan to obtain high-quality cancer data.

| PERSONALIZED MEDICINE
While great advances have been made in the management and treatment of BC over the two decades, it remains a multifaceted disease exhibiting both intertumoral and intratumoral heterogeneity as well as a variable disease course. A "one size fits all" approach is in the past and the era of PM in BC is upon us. 30,31 Empowerment is found in the ability to tell payers, providers, and patients that a drug has been specifically designed for their situation. Expanding knowledge of the molecular underpinnings comprising the etiology of cancer has driven the field of PM to identify specific tumor characteristics and develop targeted therapies against these entities. The ability to predict an individual's response to a specific therapy is the ultimate goal in modern PM. 32

| Companion diagnostics in LA
The concept of drug-diagnostic co-development, or companion diagnostics (CDx), has emerged and is now the new horizon of cancer care.
The FDA defines CDx as medical devices, often in vitro diagnostic devices that provide essential information regarding the safe and effective use of a corresponding therapy. 32,33 CDx are most often understood and applied in the context of specific molecular genomic findings that identify patients likely to respond to targeted therapy and stratify patients according to the molecular profile of the disease. 30 Although CDx testing is available in LA, 13 few studies have focused on the provision of these tests in countries that have fragmented or fragile health systems. Despite the procedural differences that may exist among the countries in this region, it is accurate to affirm that both surgical and medical oncologists request the vast majority of CDx tests in the clinical setting. In most of the Region, access to molecular testing is limited. In some countries, such as Brazil and Colombia, private and public healthcare systems only reimburse germline genetic testing ordered by certified geneticists, genetic counsellors individually, or as a part of a multidisciplinary tumor board. This imposes an additional challenge given the shortage of these specialists.

| Benefits of CDx
CDx confer numerous benefits by identifying and delivering the right drug to the right patient at the right time. Using CDx and related pharmacotherapeutic interventions can help tackle rising health expenditures by limiting drug use to those who will benefit from the drug and reducing costs related to adverse effects. Furthermore, in some cases, targeted diagnostic approaches may be more cost efficient than performing several sequential tests.
CDx are a fundamental tool for treatment decision-making with a significant impact in patient outcomes. They can benefit patient safety by reducing the frequency of invasive procedures that increase morbidity, limit frequency and severity of adverse effects, and improve compliance to treatment. CDx can improve the predictability of the oncology drug development process by allowing better selection of the target population, decreasing trials costs, and expediting timelines. [30][31][32][33] There is also added value throughout the process by providing a mechanism to collect useful data.The diverse health systems within each country, accompanied by the unique economic and regulatory stipulations require more than one financial model for CDx testing. Analyzing the vicissitudes of CDx through the lenses of all stakeholders and using investments in educational interventions are both fundamental for the consolidation of a viable model. The combination of HER2 CDx testing and trastuzumab treatment is an example of successful CDx implementation in LA. This combination is now widely covered across the Region and has evidenced significant cost benefits to healthcare systems and improved patient survival. 34

| THE ROLE OF SPECIALIZED PATHOLOGY IN BC
Selecting an appropriate diagnostic test is a challenge for clinicians, partly due to the sheer volume of choices, which is expected to increase. Additionally, contextual factors play a significant role in this decision and include treatment availability, accessibility, current standards of care, known disease characteristics, and physician experience in the use and interpretation. Collaboration between clinicians and pathologists is needed to select the right diagnostic test for right patient. Thus, pathology plays a key role in effective cancer management. However, this importance is not always reflected in the monetary investment in LA pathology departments, which often have poor infrastructure and technology, especially in the public sector.
In BC, the quality of care and clinical outcomes are directly related to the quality of breast pathology. Without accurate pathological diagnosis, clinicians may be misled, patients may receive the wrong treatment or no treatment at all, and resources may be incorrectly expended. Stakeholders must recognize breast pathology as the foundation of BC care and essential for designing any program aimed to improve the quality of care. Proper training in breast pathology for pathologists and laboratory technicians is critical and provides the underpinnings of programmatic success for any country at any level of economic wealth. 35

| MOLECULAR BIOLOGY IN CDx
The therapeutic use of molecular biomarkers relies on accurately detecting and quantifying these biomarkers for use of associated therapies. The evolution of biomarkers in BC illustrates the history of CDx and provides another perspective on how to approach current and future challenges. Although not all biomarkers are considered CDx, many function in a similar manner and provide biological, clinical, and historical context. 32 IHC is the standard method for detecting cancer biomarkers due to its ease of use, availability, use of routine microscopy, and the ability to archive stained slides. The use of both technically and clinically unvalidated IHC assays has resulted in widespread inter-laboratory variability in the preanalytical, analytical, and post analytical phases. 36 In LA, IHC validation processes are not standardized, and quality control is not common practice in laboratories. ISH techniques are only available in highly specialized laboratories and institutes. The biosimilar assay development requires a long validation process that increases the cost of the assays and prolongs the delay to test availability. protein on the cellular surface in approximately 20% to 25% of BC render HER2 an ideal actionable molecule for targeted therapy. 41,42 Trastuzumab was the first FDA-approved biologic for the treatment of HER2-positive BC and has been shown to improve survival in 50% of the neoadjuvant, adjuvant, and metastatic settings. 43,44 Due to its high prognostic and predictive value, HER2 testing is recommended for all primary, metastatic, and recurrent BCs. Currently, there are seven FDA-approved in vitro CDx devices to detect HER2 in BC, a prime example of how CDx leverages this relationship to design therapies for patients. 33 For the FDA-approved HER2 IHC CDx, different primary antibodies are used, as seen in Table 1. As with other biomarkers, HER2 IHC exhibits variability due to the inherent heterogeneity of HER2 and the inconsistencies in the processing and diagnostic phases. 45 Lessons have been learned about the IHC platform through the HER2 Herceptin franchise, which took 5 to 7 years to optimize and resulted in a $3 billion USD loss in revenue opportunity due to IHC testing problems alone. To reduce inter-laboratory variability, recommendations and workflows on HER2 IHC grading and scoring were published by the ASCO/CAP. 36 FISH, CISH, and SISH assays quantify the HER2/neu gene copy number per cell using a single-or dual-probe technique. Bright field ISH assays such as CISH and SISH eliminate the need for fluorescent microscopy, cost less, and are more stable over time. All FDAapproved ISH-based HER2 CDx employ a dual-probe technique on paraffin-embedded tumor samples. 46  . 50 Each clone is designed for a specific drug with a different interpretation and cut-off; therefore, clone-specific training is required.
Virtual pathology education helps facilitate training and should be implemented throughout the Region.

| PITFALLS IN THE DIAGNOSTIC PHASE
In LA, approximately 37% of patients are diagnosed at an advanced stage (III-IV). 51 Delays between the initial suspicion and the definitive diagnosis exist in both urban and rural areas, being greater in the latter due to the centralization of these laboratories. 52 Diagnostic decisions are further constrained by deficiencies and inequities in infrastructure, equipment, and IHC antibodies. There are limited LA studies regarding the different aspects involved in the pathologic diagnosis. Once a suspicious lesion is biopsied, the specimen goes through a process that involves pre-analytical, analytical, and post-analytical phases. However, many patients do not fully benefit from CDx testing because the specimen is unsuitable for testing due to poor quality control or improper transport. Rather than simply increasing testing access, stakeholders must examine other faults in the tumor specimen journey to ensure high quality and useful specimens for CDx testing.

| Pre-analytical phase
The pre-analytic phase, which involves cold isquemia, fixation, and pathology request forms, is possibly the most complex phase due to the difficulty in controlling the variables that impact the quality of the specimen.

| Fixation
Many studies have shown that fixation time impacts the quality of

IHC. Prolonged fixation (>72 hours) may result in discrepancies that
imply changes in the diagnostic category for HR and HER2 53 as well as FISH signal detection. 54 Conversely, shorter fixation times (<6 hours) may lead to immunoreactivity loss. To standardize the preanalytical variables, the ASCO/CAP have recommended a fixation time of 6 to 72 hours, with core needle biopsy (CNB) being on the shorter end and surgical specimens being on the longer end. 47,55 Fixation is also influenced by the concentration of fixative, pH, and buffer presence, as these factors may be responsible for masking the antigen-binding sites, altering the three-dimensional structure of proteins. 56 Therefore, the use of 10% neutral buffered formalin (NBF) is recommended. 55 In LA, the use of NBF is common practice in specialized laboratories, but not in many first level primary care hospitals, as limited commercial availability and high cost lead to using alternatives or in-house solutions. Of note, the tissue penetration rate of the fixative agent must be adequate, especially for large samples and cases requiring long transport times, for which the specimen must be sliced at 5-mm intervals and placed in sufficient NBF. 55 However, this is currently not a standard practice in all care levels.

| Cold ischemia
Cold ischemia, the time between tissue removal and fixation, is another critical step to ensure the quality of the IHC and molecular testing. The ASCO/CAP recommends a cold ischemia time <1 hour. 47 Prolonged cold ischemia times (>1 hour) may cause proteolytic degradation and have been shown to be associated with immunogenicity loss for ER/ PR, 57 HER2, 57 and PD-L1. 58 Refrigerating samples is often used in lieu of NBF in resource-limited settings but is not best practice.
Standardizing these pre-analytical variables in LA health systems is a challenge that requires coordination along the entire specimen pathway. Factors that contribute to this lack of standardization are:   are no local studies in LA. 28 Clinical utility evidence showcases the added value of a CDx to treatment management, as compared to management without. The more CDx are clinically tested, the more evidence to assure payers that these lead to overall healthcare savings.

| UNSTANDARDIZED CANCER TREATMENT
Incidence data, which is dependent on the data generated by PBCRs, are necessary for cost efficiency analyses and, ultimately, has a determining impact on public policies. 6 In LA, most decisions on healthcare resource allocation are based solely on upfront cost, which is insufficient within systems with finite resources. 60 Stakeholder perception of PM must pivot to appreciate the solutions that these innovative technologies can provide for the inequities of cancer care in LA.

| CONCLUSIONS AND RECOMMENDATIONS
Healthcare delivery to underserved populations is challenging, but inno-

GOVERNMENTS
• Develop and promote adherence to clinical guidelines in collaboration with medical societies to address fragmented care.
• Implement country-specific protocols that adapt the ASCO/CAP recommendations to national realities for the collection, handling, transportation, and processing of tumor specimens.
• Increase availability of drugs/CDx tests and address reimbursement issues by: Creating national regulatory guidelines about BC that consider various stakeholder perspectives to streamline the approval of PM strategies.
Optimizing dialogues between public and private sector stakeholders to develop country-specific sustainable funding mechanisms that support shared interests and satisfy both payer and patient reimbursement needs.

HEALTH INSTITUTIONS
• Provide a multidisciplinary approach by creating Breast Units and molecular tumor boards adapted to the constraints of resourcelimited environments.
• Improve infrastructure by Increasing investments for pathology departments to reach the level of technological and human resources required for highquality molecular testing with support of governmental funds and Establishing laboratory accreditation programs with international standards to improve quality control.
• Increase education about the best practices for handling specimens in all the personnel involved in the process.

INSTITUTIONS
• Address the shortage of medical personnel in the breast field by creating local training programs (pathology, surgical oncology, medical oncology, radiology, nursing, radiooncology) and implementing telepathology to overcome geographical disparity and strengthen second opinion networks.
• Implement education programs for healthcare professionals with three main focuses: 1. Continuous medical education (CME) (programs on BC for healthcare professionals, especially in the primary care setting 2. BC training in all specialties related to cancer care and require CME for existing specialists.
3. Awareness within BC care specialists on innovative therapies and the testing required for them.
• Increase data collection and clinical research by: Prioritizing funding and urge the importance of BC research to generate local data and define and quantify quality metrics to monitor the impact of PM.
Strengthening and creating PBCRs as part of regional and national cancer control plans Systemically measuring treatment outcomes and conduct postmarketing surveys with posterior development of costeffectiveness analysis for PM initiatives.
• Implement wide-reaching early detection strategies by

CONFLICT OF INTEREST
The co-authors declare no conflicts of interest. The organization and implementation of the consensus conference were carried out by the

DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were created or analyzed in this study.

ETHICS STATEMENT
In this review paper we follow the consideration published in PRISMA guidelines for systematic reviews and meta-analyses, step by step.