Potentially resectable
Borderline resectable
Locally advanced
Portal vein/SMV
TVI <180°
TVI ≥180° and/or reconstructable occlusion
Unable to reconstruct
Hepatic artery
No TVI
Reconstructable short-segment TVI of any degree
Unable to reconstruct
Superior mesenteric artery
No TVI
TVI < 180°
TVI ≥ 180°
Celiac trunk
No TVI
TVI < 180°
TVI ≥ 180°
When the extent of local disease exceeds the definition of BRPC above, the term LAPC is used to represent unresectable disease.
The above staging is evaluated based on diagnostic computed tomography (CT), preferably done with triple-phase contrast enhancement as part of a pancreas protocol with 3D imaging reconstruction (to allow proper visualization of the tumor–vessel interface).
Magnetic resonance imaging (MRI) and/or fludeoxyglucose positron emission tomography (FDG-PET) may also be utilized.
In addition to imaging, a pathologic diagnosis is necessary for patients who are not undergoing up-front surgical resection.
Ideally, this should be done by endoscopic-guided ultrasound with a fine needle aspiration (FNA).
A core biopsy or a cell block through multiple pass FNA is now recommended when possible in order to have adequate tissue for correlative studies when indicated [9].
If there is a biliary obstruction, placement of metal stents is preferred over plastic stents.
If patients present with clay-colored stools, flatulence and/or weight loss pancreatic enzyme supplementation should be considered.
Treatment recommendations are typically offered after baseline evaluation including a history and physical examination, complete blood count, serum chemistries, a carbohydrate antigen 19-9 (CA 19-9), and carcinoembryonic antigen (CEA).
Multidisciplinary review at a high-volume pancreas center is strongly recommended [10].
7.3 Prognostic Factors
Resectability, as determined by local tumor extension, remains the only chance for cure in the disease.
The extent of resection also carries prognostic significance—a patient who undergoes a complete (R0) resection has a better prognosis than a patient who undergoes a resection with microscopically (R1) or grossly (R2) positive margins.
Margin clearance >1.5 mm is necessary for optimal locoregional control, and the extent of clear margins in R0 resections may be used to estimate the risk of locoregional failure [11].
When PDA is resected with negative margins, the dominant predictor of prognosis is involvement of the locoregional lymph nodes at the time of resection [12].
The presence of metastatic disease at the time of diagnosis is the dominant poor prognostic indicator, with metastatic patients estimated to survive a median of 6 months.
Other prognostic factors that can help discriminate within each stage of the disease include serum and molecular biomarkers.
Common serum biomarkers include CA 19-9 and CEA [16].
A recent Australian study retrospectively analyzed the role of CEA and CA 19-9 in 393 patients treated for PDA from 2005 to 2012 [17].
Both CEA ≤6.9 and CA 19-9 ≤931.4 were positive prognostic factors for survival on univariate analysis (p < 0.001).
Significance was maintained on multivariate analysis (CEA, hazard ratio [HR] 1.27, 95 % CI 1.00–1.61, p = 0.054; CA 19–9, HR 1.38, 95 % CI 1.09–1.74, p = 0.007).
Furthermore, the linear combination of CEA and CA 19-9 (value ≤845 U/L) was one of the strongest clinicopathologic variables analyzed (HR 2.33, 95 % CI 1.84–2.96, p < 0.001).
In BRPC patients specifically, normalization of CA 19-9 (<40 U/mL after neoadjuvant therapy) is suggestive of improved survival in both resected (38 vs. 26 months, p < 0.02) and non-resected patients (15 vs. 11 months, p = 0.02) [18].
Baseline CA 19-9 greater than 90 U/mL was even reported to be a stronger predictor of death than local or distant progression in patients with LAPC treated on a phase I–II study of full-dose gemcitabine and dose-escalated intensity-modulated radiation therapy (IMRT) [19].
The number of proposed molecular biomarkers of disease severity is expanding fairly rapidly in the literature as genetic and histologic science becomes more nuanced.
Finally, the ability of each patient to successfully receive the therapies recommended by the NCCN (directed by clinicopathologic stage assessment) can also serve as a prognostic factor [20].
7.4 Molecular Biology
Advancements in the molecular understanding of PDA have made major strides over the past decade, particularly in the understanding of the genetic basis for disease.
In 2008, an initial landmark report by Jones and colleagues reported on the comprehensive genetic analysis of 24 pancreatic lesions [21].
Genetic deficiencies were identified most commonly in 12 core signaling pathways, including cell cycle control, DNA damage control, and apoptosis. Among other breakthroughs, these data also demonstrated that PDA was a cancer of relatively few mutations (averaging 63 mutations per tumor) with key genes such as KRAS, TP53, and SMAD4 frequently affected.
Currently, this information is being leveraged in research endeavors in attempts to identify a cure. Unfortunately however, the clinical phenotype of PDA has proven to be more heterogeneous than this initial work suggested.
Most recently, a genetic study in 456 tumor specimens stratified PDA into four distinct molecular subtypes that correlated with histopathologic findings: squamous, pancreatic progenitor, immunogenic, and aberrantly differentiated endocrine/exocrine [22].
With unique genetic profiles in each tumor subtype, there is a molecular basis to support a personalized therapeutic approach to the disease in the future.
Drug development, comparative clinical outcomes research, and further molecular studies will continue to shape the treatment of this disease over the next several decades.
There are several promising preclinical and early-stage clinical studies that have been initiated by an understanding of the genetic underpinnings of PDA.
These include efforts to discover novel targeted therapies and those that leverage genetic information to trial combination therapies.
One example is an interplay discovered in preclinical studies between AKT activity and radiosensitivity [23].
In early-stage translational work, the use of an oral agent enhancing the PI3-kinase/AKT pathway was associated with remarkable clinical response to radiotherapy.
7.5 Patterns of Failure
In patients with locally confined disease who successfully undergo surgical resection, there are two common patterns of failure: locoregional recurrence (in resection bed) and distant recurrence (frequently in the liver and lungs).
There are important factors that may raise the risk for locoregional disease recurrence following surgical resection, namely, lymph node-positive and margin-positive resection on histopathologic analysis. When these are present, locoregional failure occurs at a rate roughly between 50 and 70 % [24–26].
The use of radiotherapy following surgical resection, in an attempt to optimize locoregional control, remains controversial in the United States due to historical data demonstrating unclear benefit [27].
In a retrospective study of 1,130 patients with resected PDA, the patterns of failure were analyzed.
Patients in this study were divided into three groups: those who underwent surgery alone and received no adjuvant therapy (n = 392), those who received adjuvant chemotherapy after surgery (n = 291), and those who received adjuvant CRT after surgery (n = 447).
Adjuvant chemotherapy resulted in significantly fewer local and distant recurrences. This afforded an overall survival advantage (HR 0.71, 95 % CI 0.57–0.89).
Patients who underwent adjuvant CRT had fewer local recurrences but no change in distant recurrences.
Another contemporary series of 1,051 patients with resected PDA and follow-up data extending to a median of 84 months, locoregional control rates as estimated by the Kaplan–Meier method varied between 68 and 80 % [32].
The stratification between high and low rates of locoregional control appeared to be associated with the use of adjuvant CRT with the authors advocating for its routine use.
Regardless of capacity for surgical resection upon initial patient presentation, the most common cause of death is distant disease progression [33, 35].
The multidisciplinary treatment for these patients is often discussed on a case-by-case basis with options including systemic chemotherapy, palliative surgical or percutaneous interventions, and palliative radiotherapeutic approaches.
7.6 Multidisciplinary Treatment
Treatment recommendations for PDA are ideally made in a multidisciplinary setting on the basis of clinical stage, performance status, and patient preference.
As demonstrated by the Johns Hopkins experience, a multidisciplinary clinic can result in a more accurate diagnosis, and treatment recommendations can be made in a more expeditious manner [10, 38, 39].
This is achieved by successful collaboration among the numerous specialties that are dire to delivering care to patients with PDA—medical oncology, radiation oncology, surgery, radiology, pathology, palliative care and pain medicine, and gastroenterology.
For the purposes of this chapter, we will focus on trimodality care with chemotherapy, radiation therapy, and surgery.
Treatment options include:
Up-front surgery for resectable PDA
Neoadjuvant therapy for resectable PDA
Adjuvant therapy for resected PDA
Neoadjuvant/definitive therapy for borderline resectable and locally advanced, unresectable PDA
Salvage therapy for unresectable locally recurrent PDA
Systemic therapy and/or palliative therapy for metastatic disease
Systemic therapy is typically used in all stages of pancreatic cancer.
An exception to this may be patients with comorbidities that prohibit chemotherapy or poor performance status (ECOG > 2).
Goals of systemic therapy should be discussed with patients prior to initiation of therapy, and enrollment into clinical trials is strongly encouraged.
Close follow-up of patients undergoing therapy is indicated.
Neoadjuvant/definitive therapy (resectable and borderline resectable PDA):
In patients with resectable disease confined to the pancreas, up-front surgery is typically recommended; however, neoadjuvant therapy is standard at some centers [40] and is currently being evaluated in a cooperative group study (NCT01821612).
There is limited evidence to recommend specific neoadjuvant regimens for resectable PDA outside of a clinical trial, and practices vary with regard to the use of systemic therapy and radiation therapy.
Acceptable regimens include gemcitabine alone, FOLFIRINOX (FFX), or gemcitabine + albumin-bound paclitaxel (Gem/NP). Subsequent radiation (chemoradiation or stereotactic body radiation therapy, SBRT) is often recommended after maximal chemotherapy and if there is no evidence of metastatic disease.
Combination chemotherapy, particularly FOLFIRINOX and gemcitabine/nab-paclitaxel, with or without radiotherapy is beginning to repeatedly demonstrate a capacity to downstage patients into a surgical paradigm of management.
Importantly, R0 resection can be performed in this group of patients at rates exceeding 85 %.
In many of these experiences, chemotherapy is being combined with IMRT prior to resection in an effort to maximize noninvasive therapies [41].
For BRPC, a multicenter study established FOLFIRINOX followed by 5-FU-based CRT as a standard approach [42].
Induction multi-agent chemotherapy (good PFS) or single-agent chemotherapy (poor PFS) followed by RT is typically recommended.
Specifically, our institution treats with induction FFX or Gem/NP for 4–6 months followed by hypofractionated SBRT in patients with BRPC and in resectable patients who may not be able to tolerate adjuvant therapy (comorbid disease or poor performance status).
If there is direct invasion of the tumor into the bowel or stomach or there are regional lymph nodes, capecitabine-based CRT is recommended.
The goal is to select best surgical candidates and sterilize margins to facilitate a margin-negative resection and sterilization of peripancreatic lymph nodes.
More recently, SBRT has emerged as a safe and efficacious alternative to achieve local control and/or improve surgical outcomes in patients who are able to be resected.
Adjuvant therapy (resected PDA):
The use of RT after surgical resection to enhance local disease control is controversial; however, patients resected with positive margins and/or nodes should receive 6 months of adjuvant chemotherapy with or without RT.
There are both institutional and regional biases that are prevalent in the literature, with many European centers using adjuvant therapy routinely after surgery, while centers in the United States tend to be more selective [27, 43].
One area of conflict involves the lack of standardization in treatment protocols.
There are two commonly utilized external beam radiotherapy (EBRT) strategies for standard CRT: three-dimensional conformal radiation therapy (3D-CRT) and intensity-modulated radiation therapy (IMRT) delivery.
More recently, both multi-institutional analyses and pooled outcomes data suggest that there may, in fact, be a survival benefit afforded by IMRT [44, 45].
It should be noted that in these studies, similar to most in the field, selection bias is difficult to avoid, and its effects on the final results are difficult to quantify.
Despite a slight increase in the use of neoadjuvant radiotherapy between 2000 and 2010, overall use (adjuvant plus neoadjuvant) has declined over the last decade [46].
For resectable PDA, we typically recommend up-front surgery followed by adjuvant chemotherapy with or without CRT.
Locally advanced, locally recurrent, and metastatic disease:
Multidisciplinary treatment is the cornerstone of therapy for patients with PDA who present with unresectable locally advanced or local recurrent disease after resection.
Depending on performance status, mono- or combination systemic chemotherapy may be considered as initial therapy prior to radiation (CRT or SBRT) for appropriate patients with locally advanced, unresectable disease.
Patients should be evaluated for recovery from hematologic and non-hematologic toxicity prior to initiation of RT (usually 1–2-week break).
Patients who progress with metastatic disease are not candidates for RT unless required for palliative purposes.
If resected patients with a good performance status relapse (locally or distantly) after receiving adjuvant gemcitabine or 5-FU monotherapy, FFX or Gem/NP are options depending on the length of time since completion of adjuvant therapy.
If the recurrence is local only and unresectable, radiation therapy should be considered.
For patients with locally advanced/unresectable and locally recurrent PDA, we treat with induction Gem (poor PFS), Gem/NP, or FFX for 2–12 months followed by definitive CRT or SBRT.
Select patients with local obstruction or pain may benefit from up-front CRT or SBRT.
Select patients in this group may undergo attempted surgical resection after maximal neoadjuvant therapy and if technically resectable.
Clinical (radiographic) response is rare and often does not correlate with pathologic response.
Integration of MRI and/or PET/CT may assist in determining clinical response to neoadjuvant therapy.
In the setting of LAPC, SBRT has become an option in the neoadjuvant, salvage, and palliative setting with dose adjustment based on goals (Table 7.2).
Table 7.2
Literature supporting SBRT in the historically definitive, neoadjuvant, and salvage settings for localized pancreatic cancer
Study
Study design
N
Regimen
1-Year FFLP
Median OS (mos)
Acute grade ≥3 toxicity
Late grade ≥2 toxicity
Resection
R0 resection
Historically definitive setting (LAPC)
Koong et al. (2004)
(Stanford)
Prospective
6
25 Gy SBRT, 1 fraction
100 %
8.0
33 %
–
–
–
Hoyer et al. (2005)
(Denmark)
Prospective
22
45 Gy SBRT, 3 fractions
57 % (6 mo)
5.4
79 %
94 %
–
–
Schellenberg et al. (2008)
(Stanford)
Prospective
16
Gem ➔ 25 Gy SBRT, 1 fraction ➔ Gem
100 %
11.4
19 %
47 %
–
–
Chang et al. (2009)
(Stanford)
Retrospective
77a
25 Gy SBRT, 1 fraction
95 %
11.9
5 %
13 %
–
–
Mahadevan et al. (2010)
(Beth Israel)
Prospective database
36
Median 30 Gy SBRT, 3 fractions ➔ Gem
78 %
14.3
41 %
6 %
Polistina et al. (2010)
(Italy)
Prospective
23
Gem ➔ 30 Gy SBRT, 3 fractions
50 %
10.6
0
0
9 %
100 %
Schellenberg et al. (2011)
(Stanford)
Prospective
20
Gem ➔ 25 Gy SBRT, 1 fraction ➔ Gem
94 %
11.8
15 %
20 %
–
–
Mahadevan et al. (2011)
(Beth Israel)
Retrospective
39
Gem ➔ median 24 Gy SBRT, 3 fractions ➔ Gem
85 %
20
0 %
9 %
–
–
Rwigema et al. (2011) (UPMC)
Retrospective
40
Median 24 Gy SBRT
38 %
8.0b
4 %
0 %
–
–
Goyal et al. (2012)
(Case Western)
Prospective database
20
Median 25 Gy SBRT, 1 fraction
65 %
14.4
0 %
16 %
–
–
Tozzi et al. (2013)
(Italy)
Retrospective
30
Gem ➔ 45 Gy SBRT, 6 fractions
86 %
11.0
20 %
0 %
–
–
Gurka et al. (2013)
(Georgetown)
Prospective
10
Gem ➔ 25 Gy, 5 fractions
40 %
12.2
0 %
0 %
–
–
Herman et al. (2015)
(Johns Hopkins, Stanford, MSKCC)
Prospective
49
Gem ➔ median 33 Gy, 5 fractions ➔ Gem
78 %
13.9
12 %
11 %
8 %
100 %
Neoadjuvant setting (BRPC/ LAPC)
Boone et al. (2013)
(Pittsburgh)
Retrospective
25
(12 BR,
13 LA)
FOLFIRINOX +/− median 36 Gy SBRT (57 %), 3 fractions
–
–
–
8 %c
43 %
(58 % BR,
15 % LA)
33 %
(55 % BR,
10 % LA)
Chuong et al. (2013)
(Moffitt)
Retrospective
73
(57 BR,
16 LA)
Gem-based chemo ➔ median 25 Gy SBRT, 5 fractions
81 %d
16.4 BR,
15.0 LA
0 %
5 %e
44 %
(56 % BR,
0 % LA)
97 %
(97 % BR,
N/A LA)
Rajagopalan et al. (2013)
(Pittsburgh)
Retrospective
12
(7 BR,
5 LA)
Chemo ➔ median 36 Gy SBRT, 3 fractions
–
47.2
0 %
–
11 %f
92 %
(86 % BR,
100 % LA)
Mellon et al. (2015)
(Moffitt)
Retrospective
159
(110 BR,
49 LA)
Chemo ➔ median 30 Gy SBRT, 5 fractions
78 %d
18.1
(19.2 BR,
15.0 LA)
2 %
6 %e
38 %
(51 % BR,
10 % LA)
97 %
(96 % BR,
100 % LA)
Moningi et al. (2015)
(Johns Hopkins)
Retrospective
88
(14 BR,
74 LA)
Chemo ➔ median 33 Gy SBRT, 6 fractions
61 %
18.4
13.7
(14.4 BR,
18.4 LA)
3 %
6 %
22 %
(29 % BR,
20 % LA)
84 %
(84 % BR,
80 % LA)
Salvage setting (re-irradiation and/or locally recurrent)
Koong et al. (2005)
(Stanford)
Prospective
16
5-FU + 45 Gy IMRT ➔ 25 Gy SBRT boost
94 %
8.3
13 %
–
N/A
N/A
Lominska et al. (2012)
(Georgetown)
Retrospective
28
50.4 Gy IMRT ➔ median 23 Gy SBRT, 3 fractions
86 %
5.9
4 %
7 %
N/A
N/A
Wild et al. (2013)
(Johns Hopkins, Stanford)
Retrospective
18
50.4 Gy IMRT ➔ median 25 Gy, 5 fractions
62 %
8.8
0 %
6 %
N/A
N/A
Dagoglu et al. (2016)
(Beth Israel)
Retrospective
30
Median 50.4 Gy CRT ➔ 25 Gy SBRT, fractions
78 %Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access