Early detection, trial participation keys to improved outcomes in pancreatic cancer

Recent statistics suggest plenty of reason for optimism in the fight against cancer.

A report released by the American Cancer Society in January concluded the overall risk for dying from cancer has declined 20% in the past 2 decades. The death rate for colorectal cancer has fallen 46% from its peak, and similar reductions have been observed in prostate cancer mortality (45%) and breast cancer mortality (34%), the report concluded.

Yet, there are exceptions — and pancreatic cancer is the most notable.

An estimated 46,420 patients in the United States will be diagnosed with pancreatic cancer in 2014, and about 39,590 people will die of the disease that same year. Five-year survival rates have changed minimally over time, increasing from 2% in the mid-1970s to 6% in the early 21st century.

“There has definitely been an improvement in overall cancer mortality, and it is very noticeable,” Tanios S. Bekaii-Saab, MD, section chief of gastrointestinal oncology at The Ohio State University — Arthur G. James Cancer Hospital and Richard J. Solove Research Institute and a HemOnc Today Editorial Board member, said in an interview. “But when you look at pancreas cancer specifically, it’s probably the only major cancer that hasn’t seen much improvement. It remains the only cancer that is in the single digits for 5-year survival, and it’s one of the very few cancers for which mortality and incidence are pretty close to 1:1.”

Two key factors — a limited number of effective treatment options, and the fact symptoms often do not present until the disease has reached an advanced stage — contribute to the bleak outlook.

However, several experts who spoke with HemOnc Today suggest several strategies — including improved screening methods, efforts to increase clinical trial participation and a greater understanding of the genetic drivers behind certain types of pancreatic cancer — could contribute to considerable improvements in patient outcomes.

Increased awareness

Pancreatic cancer is the fourth leading cause of cancer death in the United States, but trends suggest it will become the second leading cause — behind only lung cancer — by 2020, according to the Pancreatic Cancer Action Network.

Through the efforts of the Pancreatic Cancer Action Network, a national advocacy group created to advance research and support patients, awareness of the disease is on the rise.

“A goal of the organization over the last decade has been to grow a robust pancreatic cancer scientific community,” Julie M. Fleshman, president and CEO of the Pancreatic Cancer Action Network, said in an interview.

Fleshman’s father was diagnosed with pancreatic cancer 15 years ago at age 52 years.

“I couldn’t believe the doctors were telling us nothing could be done for him,” Fleshman said. “He died 4 months later, and I came to learn that this was typical of patients diagnosed with this disease. It’s diagnosed late stage, there is no early detection method and the average survival is not good. Statistics for this year suggest 73% of patients diagnosed with pancreatic cancer will die within the first year of their diagnosis.”

Fleshman joined the organization as the first full-time employee in 2000, a year after the organization’s founding.

“When I started, we had a very small budget,” she said. “The biggest challenge in the beginning was that there really wasn’t anyone focused on studying this disease because there were no public or private research dollars being invested. In the beginning, we focused on advocating to increase federal funding that the NCI invested in pancreatic cancer in order to encourage more scientists to apply for grants and to study the disease. As our efforts grew, we were also able to fund a growing portfolio of research grants to encourage junior and senior investigators to come into this field and make pancreatic cancer their focus.”

Trial participation

The emphasis on pancreatic cancer within the scientific community has grown tremendously since then.

The Pancreatic Cancer Action Network offers a patient services program through which trained staff members help patients and their families identify specialists, learn about nutrition and palliative care, and receive information about treatment options, including clinical trials. The group has a proprietary database of current enrolling trials, and an eligibility search can match patients based on their treatment history, diagnosis and the distance they are willing to travel.

In a study published in 2013 in the Journal of Clinical Oncology, Hoos and colleagues from the Pancreatic Cancer Action Network used the database to identify the current pancreatic cancer enrollment rate and missed opportunities for improving the accrual process of pancreatic cancer trials. The organization pooled data on all open pancreatic cancer trials in the United States conducted in 2011 and 2012. Sponsors provided accrual information.

Results showed the number of open trials increased from 133 in 2011 to 167 in 2012. Most studies included patients with pancreatic ductal adenocarcinoma, and most focused on metastatic disease.

However, several challenges remain.

In 2011, the organization determined 1,804 patients — slightly more than 4.5% of patients with pancreatic cancer in the United States — had accrued to clinical trials.

Trials open in 2011 achieved an average 15% of their target enrollment, and the investigators determined trials open at that time would require a mean 6.7 years to complete accrual.

“These results suggest that harmonizing patient supply and demand for clinical trials is required to accelerate progress toward improving survival in pancreatic cancer,” the paper concluded.

The issue of trial participation is paramount to improving outcomes for patients with pancreatic cancer, Eileen M. O’Reilly, MD, a gastrointestinal medical oncologist at Memorial Sloan Kettering Cancer Center, told HemOnc Today.

“We need to express how important progress in this disease is, and the way that that will happen is by participation in trials,” O’Reilly said.

Some patients may have limited access to trials, said Andrew H. Ko, MD, associate professor in the division of hematology/oncology at the University of California, San Francisco, Helen Diller Family Comprehensive Cancer Center.

Even if they do have access, other significant barriers exist, Ko said.

“This disease is most often caught at late stages, at which point many patients have already lost a significant amount of weight and are functionally declining,” said Ko, a HemOnc Today Editorial Board member. “As a result, they are too ill or too frail to be able to consider participating in clinical trials.”

Study protocols also can be a factor. Some trials exclude patients who are taking certain drugs or have certain comorbidities, O’Reilly said. A majority only accept treatment-naïve patients.

“Trials need to match what the patients are looking for. It’s not just a demand issue, it’s a supply issue, as well,” Fleshman said.

Another barrier is patients’ limited knowledge about current trials. Hoos and colleagues determined the primary reason for low trial participation was the lack of a physician recommendation.

“Many physicians are not recommending clinical trials — they are not even having the discussion,” Fleshman said.

Biomarker detection

In January 2013, President Barack Obama signed the Recalcitrant Cancer Research Act into law. The measure requires the NCI to examine current research efforts for cancers with significantly low survival rates — including pancreatic cancer — and develop “scientific frameworks” that will lead to early detection methods and better treatment options designed to improve outcomes for these patients.

As part of that effort, the NCI issued a report in March 2014 that focused on pancreatic ductal adenocarcinoma, which accounts for 95% of all pancreatic cancer cases.

The report suggested there may be new funding opportunities to investigate the link between diabetes and pancreatic cancer. It also outlined a recently launched initiative to develop agents that target the Ras pathway, given that KRAS gene mutations are observed in the majority of pancreatic cancers.

The report also indicated there will be a program announcement for biomarkers for early detection of pancreatic cancer, as well as a commitment that the Cancer Immunotherapy Trials Network will design and conduct trials that investigate promising immunotherapies as treatment for pancreatic cancer.

The effort to establish biomarkers could be particularly valuable, Bekaii-Saab said.

“We have no good way of detecting this disease any earlier,” he said. “By the time most patients present to us, most already have metastatic disease. Even those who are lucky enough to end up in surgery, more than 90% already have metastatic disease. The disease spread is early and microscopic, we don’t see it, and nonetheless we take them to surgery with the intent of cure.”

C. Max Schmidt, MD, PhD, professor of surgery, biochemistry and molecular biology at Indiana University-Purdue University Indianapolis, said there is a need for a coordinated national pancreatic cancer early detection initiative.

“This is as simple as identifying patients at risk for pancreatic cancer and screening them,” Schmidt said. “This ‘low-hanging fruit’ should be even more emphasized since treatment strategies in patients with well-established cancer continue to be largely disappointing.”

Research conducted by Schmidt’s group at Indiana University — and published last year in the Journal of the American College of Surgeons — led to the identification of a cyst fluid protein biomarker on serous cystic neoplasms, a discovery that allows researchers to distinguish these benign cystic neoplasms from premalignant/malignant pancreatic cysts.

“Pancreatic cysts in some cases may cause pancreatic cancer, and molecular biomarkers on pancreatic cyst fluid have been discovered by our group and others to help distinguish the malignant potential of precancerous pancreatic cysts and recommend appropriate treatment,” he said. “Conversely, molecular and biochemical markers also have been discovered by our group and others, which determine which pancreatic cysts may be benign, therefore avoiding interventions in the innocent bystander.”

Mucinous pancreatic cysts have the potential to progress to invasive pancreatic adenocarcinoma, presenting an opportunity for early detection, prevention and cure, according to background information provided in the study.

Researchers hypothesized that VEGF-A levels in pancreatic fluid were associated with pathologic diagnosis. They collected pancreatic cyst fluid samples from 87 patients scheduled to undergo pancreatic resection.

Sample results indicated 24 patients had mucinous cystic neoplasm, 17 had serous cystic neoplasms, 16 had low- to moderate-grade intraductal papillary mucinous neoplasm, 11 had pancreatic ductal adenocarcinoma, 10 had high-grade/invasive intraductal papillary mucinous neoplasm and nine had pseudocyst.

When compared with all other diagnoses, VEGF-A levels were significantly increased in the fluid of serous cystic neoplasms (P<.0001). VEGF-A yielded 100% sensitivity and 97% specificity as a biomarker for serous cystic neoplasms.

“Blood tests in patients at risk for pancreatic cancer may also aid in early detection,” Schmidt said. “These include common and widely available tests: serum amylase, lipase, alkaline phosphatase, CA19-9 and HbA1c. Other novel blood tests may detect mutations in the blood of patients at risk for pancreatic cancer as an early detection marker.”

Surgery as treatment

At the Society of Surgical Oncology Cancer Symposium, held in March in Phoenix, researchers at University Hospitals Case Medical Center and Case Western Reserve University School of Medicine presented results of a study that examined the effect of surgery on survival in patients with localized pancreatic adenocarcinoma, as well as the factors that were associated with receipt of surgery.

Strohl and colleagues used the SEER database to review data on 6,742 patients diagnosed with localized pancreatic adenocarcinoma between 1988 and 2010. Researchers excluded those with an unknown site-specific surgery status, multiple primary malignancies, those who received noncancer-directed surgery, patients who underwent biopsy only, and patients with unstaged, noninvasive, regional or distant metastases at diagnosis.

Overall, 25.4% of patients diagnosed with localized pancreatic adenocarcinoma underwent surgical treatment.

Patients who were black, older than 50 years or unmarried were less likely to undergo surgery, as were those who lived in areas other than the East Coast, patients with head or body pancreatic lesions, those with higher tumor grades and patients with a tumor size greater than 2 cm (P<.0001).

Researchers observed significantly worse disease-specific survival among patients who did not undergo surgery (6 months vs. 27 months; P<.0001). Surgical treatment (HR=3.09; 95% CI, 2.82-3.38), age, sex, marital status, tumor grade and tumor size all were independently associated with survival.

Chemotherapy following surgery also clearly improves patient outcomes. One of the most encouraging trials in the past few years — the randomized, phase 3 CONKO-001 study by Oettle and colleagues — assessed the use of adjuvant gemcitabine vs. observation alone in 354 patients who underwent complete tumor resection between 1998 and 2004.

Results of the multicenter, open-label trial, were published last year in JAMA.

After median follow-up of 8 years, 87% of patients relapsed and 89.3% of patients died.

Median DFS was 13.4 months (95% CI, 11.6-15.3) among those assigned gemcitabine vs. 6.7 months (95% CI, 6.0-7.5) among those assigned observation (HR=0.55; 95% CI, 0.44-0.69).

Researchers also observed an OS benefit with gemcitabine (HR=0.76; 95% CI, 0.61-0.95). The rate of 5-year OS was 20.7% (95% CI, 14.7-26.6) in the gemcitabine arm vs. 10.4% (95% CI, 5.9-15.0) in the observation arm. Ten-year OS was 12.2% (95% CI, 7.3-17.2) in the gemcitabine arm vs. 7.7% (95% CI, 3.6-11.8) in the observation arm.

Studies examining the use of neoadjuvant chemotherapy as part of a promising and more aggressive treatment approach are scheduled to be published soon, Bekaii-Saab said.

Breaking the ‘bad spell’

Rates of pancreatic cancer have slowly increased in the past decade. The average US resident’s lifetime risk for developing pancreatic cancer is 1 in 78 (1.47%), according to the American Cancer Society.

“Part of the reason for this is that our population is aging, and the disease tends to occur more in older individuals,” Ko said.

Smoking and the obesity epidemic also may be factors, according to experts.

Despite the increased incidence, progress has been slow regarding pancreatic cancer treatments since the mid-1970s. Tumors are surrounded by dense fibrous connective tissue with few blood vessels. This makes it difficult to deliver drugs — including chemotherapy — to pancreatic tumors.

“There are a number of new regimens that have improved median survival … especially for those with advanced disease, but the improvement is small and incremental,” Ko said.

Bekaii-Saab agreed.

“When you look at the overall picture of pancreatic cancer, the only way to make improvements is to ultimately find ways to improve systemic control. We need to find regimens that control more effectively the spread of metastatic cancer,” Bekaii-Saab said. “There are studies, scheduled to begin soon, that are looking at gemcitabine combined with other therapies in the adjuvant setting. Another way to improve outcomes is to move relatively aggressive systemic therapies prior to surgery to maximize cancer control and the potential for cure.”

As with most other cancer types, targeted treatments may hold the most promise.

O’Reilly’s group is conducting research into potential targets for patients with pancreatic cancer who harbor BRCA mutations.

“It is well known that breast and ovarian cancers are associated with BRCA mutations, but also high on the list of BRCA-related malignancies is pancreas cancer,” O’Reilly said. “For this group of patients, certain drugs may have potential benefit due to ineffective DNA repair. We have an institutional registry for patients with BRCA-associated pancreas cancer and their related family members in hopes of identifying, over time, what some of the driver considerations are that make a person likely to go on to develop pancreas cancer. There won’t be definitive answers in the short term, but this is some of the research we are conducting along with trying to target the underlying genetic vulnerability of these tumors.”

Her group also is working on the development of novel agents for the treatment of metastatic pancreatic cancer in collaboration with industry partnership. Two of the drugs are OMP59R5 (OncoMed Pharmaceuticals), a notch-inhibitor stem-cell toxin, and M402 (Momenta Pharmaceuticals), a genetically engineered heparin-derivative designed to remove anticoagulant effects and to maximize the antimetastatic and antistromal effects.

“Another emerging area is the field of immunotherapy,” O’Reilly said. “We are learning in pancreas cancer that immunotherapeutic approaches may have value. I think there is much to be learned in terms of which subset of patients may benefit, and at which point in the therapeutic course value may be greatest. Treatment settings include in a postoperative setting, following standard cytoreductive therapy and possibly combined with other immune-targeting approaches or with standard therapy. There are a series of early phase 1/phase 2 trials utilizing a variety of different immune-targeting approaches, so hopefully signals will emerge over the next 6 to 12 months.”

In Schmidt’s perspective, an acknowledgement of the limitations of current therapeutic regimens may lead to more innovative, potentially more effective treatments.

“Chemotherapy in the past has often been limited to one drug — eg, 5-fluorouracil, gemcitabine, and combinations thereof,” Schmidt said. “Gemcitabine for the longest time was the only drug with reasonable toxicity to extend survival in patients with pancreatic cancer. Survival extension on average, however, was on the order of 4 to 6 weeks, and novel treatments consisted of what to add to gemcitabine to enhance its effect. Once we break this paradigm, which is beginning to happen, and admit to patients that we do not have an optimal chemotherapy treatment, we are better able to create more of a partnership with them in novel drug/drug combination discoveries through clinical trials.”

The potential for therapeutic advances has created a sense of cautious optimism in the research and clinical community.

“If you were to ask me today if anything looked promising in the pancreas cancer arena in the next year or so, I’d say no,” Bekaii-Saab said. “But in 5 years, there will be many more options for pancreas cancer patients. We are finally breaking the ‘bad spell’ that pancreas cancer has had. We are seeing these agents coming through. Things will look much better in 5 years, and we may even see this type of cancer get to where other cancer types are in terms of survival.” – by Jennifer Southall

References:

American Cancer Society. Cancer Facts and Figures 2012. Available at: www.cancer.org/acs/groups/content/@epidemiologysurveilance/documents/
document/acspc-031941.pdf. Accessed on April 15, 2014.

Hoos WA. J Clin Oncol. 2013;31:3432-3438.

Oettle H. JAMA. 2013;310:1473-1481.

Pancreatic Cancer Action Network. The alarming risk of pancreatic cancer deaths in the United States: Why we need to stem the tide today. 2012. Available at: www.pancan.org/wp-content/uploads/2013/01/incidence_report_2012.pdf. Accessed on April 15, 2014.

Siegel R. CA Cancer J Clin. 2014;64:9-29.

Strohl M. Abstract #P392. Presented at: Society of Surgical Oncology Cancer Symposium; March 12-15, 2014; Phoenix.

Yip-Schneider MT. J Am Coll Surg. 2013;218:608-617.

For more information:

Tanios S. Bekaii-Saab, MD, can be reached at The Ohio State University Comprehensive Cancer Center, A454 Starling Loving Hall, 320 W. 10th Ave., Columbus, OH 43210; email: tanios.saab@osumc.edu.

Julie M. Fleshman, JD, MBA, can be reached at Pancreatic Cancer Action Network, 1500 Rosecrans Ave., Suite 200, Manhattan Beach, CA 90266; email: jfleshman@pancan.org.

Andrew H. Ko, MD, can be reached at UCSF, Box 1705, San Francisco, CA, 94143; email: andrewko@medicine.ucsf.edu.

Eileen M. O’Reilly, MD, can be reached at Memorial Sloan Kettering Cancer Center, 1275 York Ave., New York, NY 10065; email: oreillye@mskcc.org.

C. Max Schmidt, MD, PhD, can be reached at IU Health University Hospital, Surgery OP Clinic, 550 N. University Blvd., #1295, Indianapolis, IN 46202; email: maxschmi@iupui.edu.

Disclosure: Bekaii-Saab, Fleshman, Ko, O’Reilly and Schmidt report no relevant financial disclosures. 

 

Should resection be standard for patients with localized pancreatic adenocarcinoma?

A cure for localized pancreatic adenocarcinoma cannot be achieved without surgical resection of the tumor. Therefore, surgery should remain the backbone of the therapeutic plan.

However, pancreatic adenocarcinoma is a systemic disease. This is highlighted by the fact that outcomes with surgery followed by adjuvant chemotherapy or chemoradiation — the de facto standard of care today — remain suboptimal, with median OS still less than 2 years. This points to a dire need for improving the therapeutic approach.

After many years of largely insipid results, we now have randomized controlled trials in metastatic pancreatic cancer showing marked improvement in OS with the use of multi-agent chemotherapy regimens: FOLFIRINOX (5-fluorouracil, irinotecan, oxaliplatin) and gemcitabine plus nab-paclitaxel.

We also have a growing body of literature demonstrating that pancreatic adenocarcinoma is a systemic disease from the outset. Preclinical, clinical and genomic data show that there is a high degree of genomic instability and a propensity for systemic spread even at the earliest histopathologic signs of disease, and longitudinal clinical and autopsy studies show that most patients succumb to systemic, and not locoregional, recurrence.

In addition, we can employ knowledge gained in other solid tumor malignancies. Most clinical success has been achieved in breast and colorectal cancers, with OS for localized disease approaching 90% at 5 years. This has come about with the combination of optimal local control (by refinement of surgical methods and, in some cases, the addition of radiation therapy) and effective multi-agent systemic therapy regimens. Moreover, surgical procedures for breast and colon cancers are relatively simple and well tolerated, allowing delivery of aggressive systemic therapy thereafter. When the surgical procedure is onerous — in esophagogastric or pancreatic cancers, for example — time to recovery is long with high post-operative morbidity rates, often curtailing our ability to administer adjuvant systemic therapy effectively. All these facts must be utilized to adopt a better therapeutic strategy in pancreatic adenocarcinoma.

The most promising way forward is to test multi-agent systemic therapy regimens comprising drugs with proven success in pancreatic cancer (5-fluorouracil, irinotecan, oxaliplatin, gemcitabine, nab-paclitaxel) in a preoperative setting to achieve early systemic disease control, with the ultimate objective of improving long-term OS. The component goals of this strategy are to use aggressive chemotherapy regimens (as compared with the current standard of single-agent adjuvant gemcitabine) to eradicate micrometastatic disease, to achieve systemic disease control early (gaining 2 to 3 months over the usual practice of surgery followed by adjuvant therapy), and to “select out” cases where the tumor biology is inherently resistant to chemotherapy agents, sparing them an aggressive surgical procedure that will turn out to be ineffective in a short course of time given this biology. In summary, we must tackle pancreatic adenocarcinoma as a systemic disease by testing a strategy of aggressive preoperative systemic therapy in well-designed studies to improve clinical outcomes.

References:

Boeck S. Oncology. 2007;72:314-321.

Campbell PJ. Nature. 2010;467:1109-1113.

Conroy T. N Eng J Med. 2011;364:1817-1825.

Hishinuma S. J Gastrointest Surg. 2006;10:511-518.

Rhim AD. Cell. 2012;148:349-361.

Von Hoff DD. N Eng J Med. 2013;369:1691-1703.

For more information:

Davendra P.S. Sohal, MD, MPH, is assistant professor of medicine in the department of solid tumor oncology at Taussig Cancer Institute at Cleveland Clinic. He can be reached at Cleveland Clinic, 9500 Euclid Ave., R35, Cleveland, OH 44195; email: sohald@ccf.org. Disclosure: Sohal reports no relevant financial disclosures.

We need to standardize our treatments and follow-up practices considering the most efficacious and cost-effective practices.

Many patients with early-stage disease are not offered and/or do not undergo surgery, and we know national assessments of operative staging and lymph node evaluations after surgery are inadequate.

There have been some recent successes with combination therapy for advanced pancreatic cancer — combination chemotherapy with FOLFIRINOX or gemcitabine plus nab-paclitaxel have resulted in improvement in survival by as much as 2 to 4 months. However, many patients are not candidates for the multidrug regimens, either because of poor performance status or poor end-organ function.

Moreover, targeted agents thus far have had limited success. Recent trials adding targeted therapies have either failed to improve response to chemotherapy (eg, in the case of the hedgehog inhibitor IPI-926 [Infinity Pharmaceuticals]) or have been found to have unexpected toxicity (eg, in the case of pegylated hyaluronidase PEGPH20 [Halozyme Therapeutics]). Furthermore, given the extent of the number of mutations and the epigenetic changes associated with pancreatic cancer progression, targeting one or two pathways may not be sufficient to improve outcome in the vast majority of patients with advanced disease.

Current studies have shown that we also must continue to promote high-quality cancer care by doing the things we know how to do. Compared with a 5-year survival of about 5% in patients with metastatic disease, patients with early-stage pancreatic cancer (stage I) have a 5-year survival of about 30%. Improved staging is a prerequisite for the success of future adjuvant therapy trials. We must work harder to have every patient with pancreatic cancer care evaluated by a multidisciplinary team and reviewed in a tumor board setting. We need to standardize our treatments and follow-up practices considering the most efficacious and cost-effective practices.

Unfortunately, without any effective measure of screening for early-stage pancreatic cancer, most patients will continue to be diagnosed with advanced disease and our current therapies at this stage are largely non-curative. We need to do a better job of communicating this — and the financial implications associated with treatment — to our patients.

References:

Bernard DS. J Clin Oncol. 2011;29:2821-2826.

Bilimoria KY. Cancer. 2007;110:1227-1234.

Bilimoria KY. Ann Surg. 2007;246:173-180.

Bilimoria KY. Arch Surg. 2008;143:671-678.

Bilimoria KY. J Natl Cancer Inst. 2009;101:848-859.

Commission on Cancer. Cancer Program Standards 2012: Ensuring patient-centered care. 2012. Available at: www.facs.org/cancer/coc/programstandards2012.pdf. Accessed on April 24, 2014.

Conroy T. N Engl J Med. 2011;364:1817-1825.

Halozyme. Halozyme Announces Clinical Hold of PEGPH20 Pancreatic Cancer Trial Following Voluntary Halt of Trial by Halozyme. 2014. Available at: www.halozyme.com/Investors/News-Releases/News-Release-Details/2014/Halozyme-Announces-Clinical-Hold-of-PEGPH20-Pancreatic-Cancer-Trial-Following-Voluntary-Halt-of-Trial-by-Halozyme/default.aspx. Accessed on April 24, 2014.

Hu YY. J Natl Cancer Inst. 2012;104:1164-1172.

Katlic MR. CA Cancer J Clin. 2011;61:382-396.

Kindler HL. J Clin Oncol. 2010;28:3617-3622.

Lang M. Boston Business Journal. Infinity Pharma halts pancreatic cancer trial. 2012. Available at: www.bizjournals.com/boston/blog/mass-high-tech/2012/01/infinity-pharma-halts-pancreatic-cancer-trial.html. Accessed on April 24, 2014.

Leighl NB. J Clin Oncol. 2011;29:2077-2084.

Malin JL. J Clin Oncol. 2010;28:3212-3214.

Merkow RP. Ann Surg Oncol. 2014;21:1067-1074.

Shankaran V. J Clin Oncol. 2012;30:1608-1614.

Tomlinson JS. Arch Surg. 2007;142:767-773.

Von Hoff DD. N Engl J Med. 2013;369:1691-1703.

Weeks JC. N Engl J Med. 2012;367:1616-1625.

Wright FC. Ann Surg Oncol. 2011;18:629-635.

For more information:

David J. Bentrem, MD, MSCI, and Hidayatullah G. Munshi, MD, are associate professors in the division of hematology/oncology at Northwestern University's Feinberg School of Medicine and Robert H. Lurie Comprehensive Cancer Center. They can be reached at Northwestern University Feinberg School of Medicine, 420 E. Superior St., Chicago, IL 60611; email: dbentrem@nmff.org and h-munshi@northwestern.edu. Disclosure: Bentrem and Munshi report no relevant financial disclosures.