|
|
Back to 2016 Annual Meeting
Evolution of Outcomes of Patients With Pancreatic Ductal Adenocarcinoma Undergoing Pancreatoduodenectomy following Preoperative Therapy
Jordan M. Cloyd*1, Matthew H. Katz1, Jason B. Fleming1, Robert A. Wolff2, Gauri R. Varadhachary2, Milind Javle2, Rachna T. Shroff2, David Fogelman2, Michael Overman2, Christopher Crane3, Eugene H. Koay3, Bruce D. Minsky3, Prajnan Das3, Douglas B. Evans4, Jeffrey E. Lee1
1Surgical Oncology, MD Anderson Cancer Center, Houston, TX; 2Medical Oncology, MD Anderson Cancer Center, Houston, TX; 3Radiation Oncology, MD Anderson Cancer Center, Houston, TX; 4Surgery, Medical College of Wisconsin, Milwaukee, WI
Background: The administration of chemotherapy and/or chemoradiation prior to pancreatectomy for pancreatic ductal adenocarcinoma (PDAC) is increasingly reported, especially for patients with borderline resectable (BR) disease. We sought to both document changes in treatment strategies and evaluate outcomes of patients who underwent pancreatoduodenectomy (PD) following preoperative therapy over a 25-year period.
Methods: 599 consecutive patients with PDAC who underwent PD at a single high-volume institution following the receipt of chemotherapy and/or chemoradiation between 1990 and 2014 were retrospectively reviewed. Components of multimodality therapy, clinicopathologic characteristics and long-term outcomes in four successive time periods (1990-1999, 2000-2004, 2005-2009, 2010-2014) were evaluated and compared.
Results: The average number of patients per year who underwent PD following preoperative therapy increased over time (15, 27, 30, 34, p<0.001). The proportion of operations performed for radiographically resectable tumors decreased, while operations for BR and locally advanced (LA) cancers increased (Table). As expected, the preoperative regimens used evolved significantly over the study period (Table). Statistically significant increases were observed in the need for vascular resection and R0 margin rates, while mean EBL decreased and the rates of pathologic complete responses remained relatively unchanged. Despite the increase in proportion of patients with BR/LA disease, locoregional recurrence rates remained unchanged over time (16.1%, 26.1%, 20.9%, 19.6%, p>0.05). In contrast, overall survival (OS) improved significantly (median 24.1mo, 28.3mo, 38.0mo, 38.1mo; Figure). On Cox proportional hazards regression, time period was associated with improved OS in a step-wise fashion (1990-1999: reference; 2000-2004: OR 0.83, 95% CI 0.64-1.08; 2005-2009: OR 0.70, 95% CI 0.55-0.90; 2010-2014: OR 0.59, 95% CI 0.42-0.83).
Conclusions: The data herein, which reflect changes in referral patterns, refinement of disease staging algorithms, expansion of the role of pancreatectomy and an evolving preoperative treatment strategy, identify a modest but distinct improvement in the outcomes of patients who underwent PD following preoperative therapy over time. Further improvement will require disruptive advances in preoperative regimens in addition to continued attention to meticulous surgical care.
| | 1990-1999 (149) | 2000-2004 (134) | 2005-2009 (148) | 2010-2014 (168) | p-Value | Radiographic Staging
Potentially Resectable (%)
Borderline Resectable (%)
Locally Advanced (%)
Unknown (%) | 124 (83.2)
2 (1.3)
7 (4.7)
16 (10.7) | 117 (87.3)
10 (7.5)
7 (5.2)
0 | 106 (71.6)
33 (22.3)
9 (6.1)
0 | 102 (60.7)
52 (31.0)
14 (8.3)
0 | <0.0001 | | CA 19-9, median | 123.0 | 159.5 | 108.0 | 128.1 | 0.7 | Preoperative Chemotherapy
None
Gem+Other
FFX or Gem+Abr
Other | 147 (98.7)
0
0
2 (1.3) | 65 (48.5)
63 (47.0)
0
6 (4.5) | 51 (34.5)
80 (54.1)
0
17 (11.5) | 61 (36.3)
47 (28.0)
46 (27.4)
14 (8.3) | <0.0001 | Preoperative Radiation
Hypofractionated
Standard Fractionated
None | 102 (68.5)
41 (27.5)
1 (0.7) | 110 (82.1)
14 (10.4)
5 (3.7) | 46 (31.1)
92 (62.2)
4 (2.7) | 39 (23.2)
112 (66.7)
17 (10.1) | <0.0001 | Operative
EBL, mL, mean (SD)
Vascular Resection (%)
LNs Excised, mean (SD) | 1563 (1200)
62 (41.6)
16 (8.4) | 880 (793)
46 (34.3)
21 (7.3) | 813 (729)
57 (38.5)
25 (8.7) | 750 (600)
84 (50.0)
29.3 (11.7) | <0.0001
0.04
<0.0001 | Pathology
Tumor Size, cm, mean (SD)
R0 Margin Status (%)
Positive LNs (%)
Pathologic Complete Response (%) | 3.0 (1.4)
125 (83.9)
73 (49.0)
4 (2.7) | 3.0 (1.4)
120 (89.6)
84 (62.7)
4 (3.0) | 2.0 (1.2)
140 (94.6)
75 (50.7)
2 (1.4) | 2.4 (1.4)
157 (93.5)
86 (51.2)
8 (4.8) | 0.06
0.01
0.09
0.2 |
Gem, Gemcitabine; FFX, folfirinox; Abr, Abraxane; EBL, estimated blood loss; SD, standard deviation; LN, lymph node 
Back to 2016 Annual Meeting
|
