mFFX—no 5FU bolus—4 cycles, then RT/cape; gemcit postop
Accrual rate, toxicity, CR/PR, completion of all therapy, R0/R1
Medical University of South Carolina
mFFX—no 5FU bolus—6 cycles then RT/cape
R0/R1 resection, (OS, TTR, ORR, path CR) and safety
University of Maryland
mFFX—no 5FU bolus—4 cycles then SBRT
Resectability, DFS, OS, TTR, path CR, and safety
Standard full dose FFX
Assess safety and efficacy (OS, PFS, ORR)
Foundation for Liver Research/Erasmus Medical Center
Standard full dose FFX—4 cycles then SBRT
OS, radiologic RR, resection rate, PFS, biologic predictive markers
Massachusetts General Hospital/NCI
Standard full dose FFX—8 cycles plus losartan then proton beam RT
Feasibility, PFS, OS, toxicity, downstaging, gene mutations
In the last few years, there have been definite indications that supplementary therapy with more targeted agents, such as the EGFR inhibitor erlotinib , and genotype-directed choices such as PARP (poly(ADP-ribose) polymerase) inhibition with olaparib in BRCA-1 and 2 mutated individuals  could be the way of the future. The decision to use gemcitabine based on expression of the nucleoside transporter HENT-1 [32, 33] is also being investigated along with many other initiatives such as targeting the KRAS pathway  and the tumoral stroma , and these results are eagerly awaited. Finally, immunotherapy may shortly become yet one more therapeutic option . Vaccines and immunomodulating agents are undergoing rigorous development and study and hold considerable promise, with a recent report of GVAX Pancreas Prime and Listeria Monocytogenes-Expressing Mesothelin (CRS-207) Boost Vaccines showing considerable prolongation of overall survival (9.7 versus 4.6 months) in patients with metastatic disease .
Imaging and Pathology
In parallel to the evolution in treatment of pancreatic cancer, there has been equally impressive progress in the imaging of this disease and in the ability to radiographically characterize the stage and biology of individual tumors. Understanding the technical developments in these areas underlines the importance of effective multidisciplinary collaboration. Thin slice multi-detector CT imaging with oral and intravenous contrast, using non-contrast, arterial, pancreatic parenchymal, and portal venous phases, is critical to accurate assessment of the tumor and blood vessels. Anything less than the full pancreatic protocol sequences results in suboptimal imaging [38, 39]. If the patient has a contrast allergy, renal insufficiency, is pregnant or has inconclusive results on CT, then a pancreas protocol MRI may be used. MRI can be of particular value if small metastases to the liver or peritoneum are suspected . Diffusion weighted MRI may even be of value in predicting response to neoadjuvant therapy . Positron emission tomography (PET) scanning is increasingly being used to assist in planning for radiation therapy and to detect early metastatic disease not seen on routine imaging [42, 43]. It may also distinguish benign peripancreatic inflammation from metastases or local extension of malignancy .
Accurate staging is particularly germane in the management of patients with borderline resectable and locally advanced pancreatic cancer where agreement on definition is critical and standards of care are just now being developed to enable standardization of the therapeutic approach. The Society of Abdominal Radiology and the American Pancreatic Association have constructed a radiology reporting template which ensures that there is high quality and reproducibility to the reports, and this approach should rapidly be adopted by all institutions . An abbreviated version of this template, as pertains to the critical vascular structures, is illustrated in Table 5.2.
Abbreviated radiology checklist
[Include degrees (0–360) and length (mm/cm) of involvement or occlusion with image numbers]
1. Superior mesenteric artery (SMA): [no evidence of involvement/involvement]
2. Superior mesenteric vein (SMV): [no evidence of involvement/involvement]
3. Portal vein: [no evidence of involvement/involvement]
4. Celiac axis: [no evidence of involvement/involvement]
5. Hepatic arteries (common, proper, right, and left): [no evidence of involvement/involvement]
6. Gastroduodenal artery (GDA): [no evidence of involvement/involvement]
7. IVC and aorta: [no evidence of involvement/involvement]
1. First jejunal branch of SMV: (anterior or posterior) to SMA
2. Variant anatomy
Following surgery, the College of American Pathologists recommends a comprehensive and standardized analysis of each resected tumor which includes size, margins, histologic grade, nodal involvement, vascular involvement, lymphovascular invasion, perineural invasion, intraepithelial neoplasia, and any evidence of chronic pancreatitis . This has permitted a more accurate assessment of the cancer prior to and following any therapeutic intervention and a more effective selection of appropriate local and systemic therapy. Handling of specimens, margin analysis, and the definition of an R1 resection still differs among pathologists in the USA and Europe. These variations need to be considered when analyzing study results . An assessment of response to neoadjuvant therapy by the group at MDACC has determined that a full pathologic complete remission (2.7 %) or presence of minimal residual disease (16.1 %) following therapy has a much better prognosis than a moderate (55.6 %) or minimal response (25.6 %) and correlates with better survival .
Endoscopy and Interventional Radiology
In the last 5 years, we have seen remarkable advances in endoscopic and interventional radiologic capabilities. Upper endoscopy , ERCP, and EUS have made a significant impact on the safety and accuracy of diagnosis and staging and on the management of common problems such as pain (celiac axis block) , biliary obstruction , duodenal obstruction, and postoperative complications. Duodenal obstruction has become an increasingly common problem in locally advanced disease as modern therapy has extended survival, with one recent series reporting an incidence of 38 % . Treating duodenal obstruction has become relatively routine, but managing the concomitant biliary obstruction when that occurs, frequently requires multidisciplinary experience. Creative endoluminal techniques such as double stenting and transmural biliary drainage techniques may be needed [51, 52].
Advances in interventional radiology have greatly contributed to improved management of symptoms related to disease and thus have allowed patients to initiate and complete neoadjuvant therapy. Percutaneous approaches to celiac neurolysis have improved the management of epigastric and back pain associated with celiac plexus infiltration , and percutaneous endovascular stenting has alleviated symptoms of intestinal and hepatic ischemia from tumor invasion into the SMA or hepatic arteries . Percutaneous transhepatic portovenous stenting has been shown to be effective in the management of ascites resulting from portovenous tumor thrombus . Finally, splenic artery embolization for non-operative management of low platelet counts as a result of hypersplenism has allowed continuation of dose-intensive chemotherapy in cases in which this has been important .
Key Components of the Multidisciplinary Approach
All these developments have created a menu of therapeutic and palliative treatment options that, individually, have uncertain potential benefit for any given patient. Each of the options in this menu has to be duly considered, with the risks and benefits carefully weighed. The optimal way to do so is to have every patient evaluated by individual experts from each of the disciplines involved, for there to be discussion among the members of that team, and a consensus approach to care developed.
Abundant evidence exists to support the notion that multimodality management of pancreatic cancer is associated with improved outcomes, with some centers demonstrating 5-year survival rates as high as 27 % among patients treated with multimodality therapy compared to 10–15 % in historical series with surgery alone . There is also, however, considerable evidence that compliance with established national guidelines, such as that of the NCCN , remains disappointingly low , that there is considerable variability in the quality of the treatment of pancreatic cancer in the USA and that multimodality care is, as yet, still quite uncommon . The optimal way to foster this multimodality care bears some scrutiny as different approaches may work better in different settings, and various disease specific groups may learn from one another. As the complexity of oncologic care continues to increase, it will be more important than ever that all involved in the care of the patient be experienced, knowledgeable, and current on new developments and that these experts communicate effectively.
The Multidisciplinary Conference
The cornerstone of quality multimodality care is most often the multidisciplinary conference (tumor board, cancer conference, HepatoPancreatoBiliary conference) which may meet on a weekly, biweekly, or monthly schedule. This gathering has been indispensable at well-developed cancer centers in regularly bringing all members of the care team together in a single forum in which each case can be reviewed and in which plans for treatment and research can be tailored to the specific needs of the patients and their pathology. The importance of multidisciplinary discussion is perhaps best exemplified by the finding, in a 2007 review of national practice from the national cancer database, that up to 40 % of patients with resectable peri-ampullary cancers were not offered surgery . This finding was directly attributed by the authors to nihilistic bias on the part of certain members in the care team and would seem to reflect a lack of communication between treating and referring physicians.
The multidisciplinary conference as an entity has been studied. It has been established that the decision-making process promoted by these meetings reduces the variability engendered by physicians acting independently , promotes enrollment on study protocols , and is essential for the integration of clinical information into quality biospecimen repositories . It has been well documented that the conferences have a definite impact on the ultimate care plan, with multiple studies demonstrating that the ultimate recommendations for therapy are frequently changed (up to 43 % of cases) by the consensus opinion developed in the conference . One recent prospective evaluation of practice patterns at a large tertiary cancer center found that 84 % of physicians were somewhat or very certain of their plans prior to conference and still changed their plans in 36 % of cases (72 % of those changes qualified as major changes) based on the conference’s consensus recommendations . The recognized importance of these meetings is underlined by the fact that both the Commission on Cancer and the American College of Surgeons require that institutions seeking accreditation have multidisciplinary conferences prospectively reviewing cases and discussing management decisions (Cancer Program Standards/American College of Surgeons). It is clear that institutional efforts to ensure accurate pathologic staging via synoptic analysis, to develop standardized templates for radiologic reporting, and to standardize protocols for therapy, promote cost-effective care that provides the best outcomes for the patients .
While there is a reasonably common format at larger institutions, it is worth reviewing the essential elements of an effective tumor board [61, 64, 65]. These elements are tabulated in a checklist format in Table 5.3. Meetings are increasingly organized by cancer type as treating physicians become more subspecialized. The conferences should ideally take place on a weekly schedule, thus enabling timely discussion and disposition of cases. While this frequency may not be feasible at all institutions, it promotes timely treatment planning and minimizes the time a patient waits for a decision regarding formulated algorithms. Appropriate physical space needs to be available on a regularly scheduled basis. This includes adequate seating for all participants, confidentiality (HIPPA), and availability of the necessary equipment for projection of radiologic images, endoscopic images, and microscope slides (and also videoconferencing if needed). Ideally, audiovisual/IT equipment may also permit the projection of computerized data such as treatment schemas, standards of care, investigational protocols, genomic and proteomic analysis, and collated data. Suggested participants should include all of the following: medical oncologists, surgical oncologists, radiation oncologists, gastroenterologists with endoscopic expertise, diagnostic and interventional radiologists, pathologists, geneticists, and a tumor registrar. We would argue that, under optimal conditions, having more than one individual from each discipline at the meeting affords more effective evaluation of the available treatment options. In reality, it is often difficult to consistently have more than one member of a given discipline at the conference. At a minimum, it is desirable to have multiple representatives from surgical oncology and medical oncology present. Additional members of the board may include oncology nurses, social workers, palliative care physicians and staff, nutritional services, pastoral care, and the patients’ primary care physicians. Fellows, residents, medical students, and other trainees should attend and be encouraged to participate by case presentations and other means .
Comprehensive multidisciplinary conference checklist
Clearly designated leader
Weekly schedule for timely discussion and disposition of all cases
Appropriate physical space—adequate seating, confidential (HIPPA), quiet
Appropriate equipment—projecting microscope, IT/visual equipment for projection of radiology and endoscopy images
Audiovisual equipment for virtual meetings if needed
Appropriate representation by all specialties: surgery, medical oncology, radiation oncology, interventional radiology, gastroenterology, radiology, pathology, primary care
Additional desirable staff: nurses, social workers, nutritionists, pastoral care, geneticists, tumor registrar, and research associates
Fellows, residents, and students where applicable
Documentation of discussion and recommendations in secure, retrievable location
Continuing Medical Education credits
Method for communication of results to all stakeholders not present
Attendance at these meetings by more marginal participants may be better when some form of incentive, such as continuing medical education (CME) credit, is provided. This is likely to be dependent on location and culture, and does not appear to be as important at academic centers (where many alternative sources of CME are available). One recent study of MDC conferences at a single academically affiliated tertiary center reported that only 24 % of attendees sought CME credits for their participation . The time spent in such activity, in lieu of actual patient care, should be recognized by the institution, and participants should not be censured in any way. A clearly designated leader is essential to the smooth functioning of the conference [68, 69]. This individual should have the personal and professional respect of the members of the board, be knowledgeable and experienced, and be able to foster appropriate discussion. He/she should maintain decorum and ensure that differing opinions are rightly heard, that the meeting moves forward, and that it does not get bogged down over discussion of any one case. Finally, documentation of all cases presented in a concise, accessible format, along with diagnostic or therapeutic recommendations allows those not present to easily access this information. This facilitates the retrieval of data needed for analysis of conference utilization, tumor volumes, trends, and tissue banks.
Importantly, it should be noted that recommendations from a tumor board are recommendations and are not legally binding. These recommendations do not relieve the treating physician from the obligation to provide care for the patient. The treating physician must critically scrutinize the recommendations before implementation and, ideally, any deviation from these recommendations should be clearly explained based on the obligation to treat the patient safely and effectively .
Alternative and Complementary Arrangements
Several other elements of a multidisciplinary approach can be valuable adjuncts to the cancer conference but are not as commonly recognized and not as often employed across the country. A prime example is the multidisciplinary clinic . For many reasons, this has not been as widely adopted as the multidisciplinary conference but may be of equal or greater value. Challenges to establishing this arrangement have included: limited clinical space to accommodate a larger group at one time; arranging appropriate support staff; scheduling of sequential patient visits; conflicting needs of surgical, oncologic, and medical specialties; entrenched attitudes to clinical care; and billing for services. Despite the logistic difficulties in establishing and maintaining these clinics, they have been consistently identified as enhancing the efficiency of care by allowing patients to see all of their care team in one visit. The clinics provide the opportunity for real-time interaction between members of the team who are treating diseases for which conditions change frequently. Including a specialist clinical cancer pharmacist in the clinic results in improved medication adherence (p = 0.007) and patient satisfaction (p < 0.001) .
In one notable study of the efficacy of a pancreatic cancer multidisciplinary clinic at Johns Hopkins Hospital, 25 % of patients had their care plan revised after analysis in the clinic, with both upstaging (29/38 patients) and downstaging (9/38 patients) of the original classification of the extent of disease . Radiology review contributed the most to a change in plans (18.7 %) and pathology review was also important (3.4 %). One notable change in care was the determination of resectability in those cases where the portal vein/superior mesenteric vein confluence was involved. Patients identified in this study as having tumors involving the portal/SMV confluence had frequently been evaluated by programs at smaller referring hospitals, been deemed to have unresectable disease at these institutions and then were reclassified as resectable or borderline resectable in the multidisciplinary clinic. Registration into the National Familial Pancreatic Cancer Tumor Registry was noted to increase from 49.2 to 77.8 % following initiation of this clinic, and participation in clinical studies was offered to 51/203 patients. The clinics also promote the academic mission, both by creating the appropriate environment to teach students, residents, and fellows in a multidisciplinary setting emblematic of modern oncology and by facilitating the determination of patient eligibility for clinical studies .
Virtual tumor boards , with secure access to protect patient confidentiality, are increasingly prevalent in regions where clinical volumes and practice patterns make a regular multidisciplinary conference practically impossible. This is particularly useful in a rural setting and where a large institution may have affiliates with which it wishes to coordinate care and clinical study accrual [74, 75]. This can also be helpful in complex diseases such as pancreatic cancer and in which access to tertiary care may be essential for multidisciplinary management and in which triaging patients for rapid referral may be critical . In-person meetings where possible, however, still remain preferable as interactions are easier, the discussions are less regimented, and the number of cases presented is often greater .
Evolving experience with personalized care now suggests that a molecular or genomics tumor board may be a necessary addition to the more standard cancer conference discussed above, and many larger centers have such an entity. Most physicians do not have formal training in the evaluation of advanced genomics, and basic scientists, geneticists, and experts in bioinformatics may all contribute to the interpretation of results. In a recent series, 34 patients presented at a university molecular tumor board had a median of 4 molecular abnormalities each on next-generation sequencing, and no two patients had the same profile . Eleven of 34 patients had treatment decisions informed by this test, and three of 11 had a meaningful response to treatment that was determined by molecular or genomic profiling . Barriers to therapy in those not treated were mainly related to access to appropriate agents. As the cost of genome sequencing declines, more and more patients with pancreatic cancer will have their cancers tested. Given the limited efficacy of current chemotherapy in pancreatic cancer, any leads engendered by this approach will be eagerly investigated.
Additional Participants and Resources
An often unsung member of the multimodality care team, not typically included in multidisciplinary conferences and clinics, is the primary care physician. Almost all patients with pancreatic cancer have concomitant medical conditions which require ongoing care and which are not optimally managed by physicians in specialty settings. These conditions may include diabetes mellitus, hypertension, malnutrition, cardiovascular disease, thrombosis and embolism, and intractable pain . Further, the often long-standing relationship between patient and primary MD is a source of great comfort to many individuals, assuring them that they are following the right path in the treatment of their disease and providing essential moral support [80, 81]. Communication between the primary MD and the specialists is often not optimal but can and should be improved with better organization and structure . Equally, all physicians are supported by nurses, technicians, dieticians, social workers, research associates, pharmacy staff, phlebotomists, and front office personnel. All of these individuals provide substantive psychosocial support for patients and technical skill sets without which their medical care would not be possible. These caregivers, and others, need to be rightly recognized as essential members of the cancer care team.
Finally, it is important to recognize the family and friends of the patient. These individuals are critical to the success of both the simplest and most complex treatment plan and without question impact clinical outcomes. Patients often rely on family members to provide transportation to treatment centers, fill prescriptions, administer medications, call insurance companies, report complications to treating physicians, and manage households, in addition to offering emotional support and encouragement . Because of the often terminal nature of this disease and the rapidity with which it may progress, there is frequently little time for caregivers to adjust to the circumstances facing their loved ones [84, 85]. These individuals may also be concerned about their own genetic risk and that of other family members . It is becoming increasingly clear that although patients with well-developed support infrastructure and healthy caregivers cope better with the stress and the complexity of treatment than those who do not have such a system in place, the caregivers may themselves be at risk of illness and mortality . As the disease progresses, these issues intensify and the quality of life of the caregivers may deteriorate significantly such that psychiatric care may be needed . This is clearly a neglected aspect of comprehensive pancreatic cancer care, and future studies to examine more effective and meaningful interventions on their behalf are sorely needed. A recent pilot study to assess the experience of caregivers has demonstrated that these data are not only needed, but that the caregivers are very willing to share their stories and to seek assistance wherever they may find it .
Impact of Multimodality Care on Sequencing of Therapy
With the above infrastructure in place, a given case can be duly considered and the critical decisions necessary for optimal therapy can be made. As this decision-making takes place, a number of issues are critical. First, the goals of therapy must be clearly delineated—cure versus palliation versus other. It may not always be possible to do this, especially in borderline resectable disease where ultimate curability is uncertain, but a thorough understanding of the status quo on the part of the patient is essential if his/her expectations regarding the ultimate outcome are to be realistic. Second, the determination of eligibility for clinical studies is vital if important progress is to be made in this disease. It is estimated that only a very small fraction of all eligible patients are enrolled in clinical studies (3 %), with underserved populations rarely, if ever, exposed to available clinical studies and particularly prone to being neglected in this regard . Third, individualized plans are ideally made with consideration given to stage of disease, tumor molecular profile, inheritance patterns, comorbidities, performance status, fragility scores, and cultural issues. Fourth, the skill set of the treating physicians and institutional experience must be considered. Many recent studies have demonstrated that outcomes of complex surgical procedures are better when the procedures are done in large centers performing a critical number of procedures on an annual basis . When centers and physicians attempt procedures and therapies that they are not equipped or experienced enough to effectively carry forward, patients will not infrequently then need to be referred to tertiary centers where they may be offered second operations or attempts at treatment. Second procedures frequently do not have the desired results and an opportunity to cure may be lost . Despite existing data supporting the importance of experience, there does not appear to be a migration to high-volume centers. There is a perception on the part of patients that remaining close to home is of definite benefit, and many procedures continue to happen in less experienced settings .