Progress in Pancreatic Cancer: ASCO Studies Move the Needle

Richard L. Schilsky, MD

New strategies for the use of standard chemotherapies are likely to improve outcomes for a significant subset of patients with pancreatic cancer, according to study results presented at this year's ASCO Annual Meeting (ASCO 2018).

The findings are generating excitement in the field at a time when pancreatic cancer remains among the most deadly and intractable malignancies.

PRODIGE 24

PREOPANC-1

PRODIGE 35/PANOPTIMOX

ASCO 2018 featured the results of 3 studies that are expected to influence the way therapy is administered in adjuvant, neoadjuvant, and metastatic settings:The study found that a 4-drug chemotherapy combination dramatically improved median overall survival (OS) to 54.4 months compared with 35.0 months for standard gemcitabine in a selective population of fit patients who had undergone resection.¹Preliminary data showed that perioperative chemoradiotherapy resulted in a median OS of 17.1 months versus 13.7 months with immediate surgery and adjuvant chemotherapy in a broader population of resectable or borderline resectable patients.²A "stop-and-go" strategy involving a less intensive chemotherapy regimen followed by reintroduction of standard agents upon progression proved to be a viable maintenance strategy as frontline treatment for patients with metastatic pancreatic cancer.³

The adjuvant therapy data from the PRODIGE 24 trial drew the most enthusiastic responses. “This is probably the biggest advance we’ve seen in pancreatic cancer since the introduction of gemcitabine, which is now about 25 years ago," Richard L. Schilsky, MD, ASCO's chief medical officer, said in a discussion about the results of at a press conference. "This is a huge step forward, at least for those patients who are able to undergo surgical resection."

The findings are “pretty impressive,” Tanios S. Bekaii- Saab, MD, senior associate consultant at Mayo Clinic in Phoenix, Arizona, said in an interview with OncLive^® NewsNetwork. “This is a disease that just a few short years ago, when we were looking at the adjuvant data, what we would hope for is 22 or 23 months of survival. This is doubling survival...It moves the needle and moves it quite nicely; it should be considered a standard for those who are considered eligible.”

Bekaii-Saab expressed optimism about study results that are being reported throughout the gastrointestinal (GI) cancer landscape, including pancreatic cancer. “Overall we’re seeing a lot of exciting data. We are on the right track and asking the right questions,” he said. “One thing that’s lacking but is being worked on very aggressively is starting to break down these GI cancers into multiple buckets—molecularly driven, immune-driven buckets—and start piece-mealing these cancers into these buckets.”

Combination for Adjuvant Setting

The statistics for pancreatic cancer show that advancements are urgently needed. Pancreatic cancer is expected to account for 3.2% of new cancer diagnoses in the United States in 2018 but is the third-leading cause of cancer death, according to analyses from the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute. Fifty-two percent of the cases are discovered after the cancer has metastasized and, although the 5-year relative survival rate has improved since 2000, it is estimated at 8.5%.⁴In the phase III PRODIGE 24 study, investigators from Unicancer, a network of comprehensive cancer centers in France and a leading clinical trial sponsor in Europe, compared a modified FOLFIRINOX regimen (mFOLFIRINOX) with gemcitabine in patients with nonmetastatic pancreatic ductal adenocarcinoma treated at 77 French and Canadian centers from April 2012 to October 2016.

Eligible patients had undergone surgery and had no tumor residue on a postoperative computed tomography scan. Participants had to be between the ages of 18 and 79 years with World Health Organization performance status scores of ≤1. Additionally, they were required to have a carbohydrate antigen (CA) 19-9 radio-immunoassay level of <180 μ/mL within 12 weeks after surgery.

Three to 12 weeks after surgery, participants were randomized 1:1 to receive mFOLFIRINOX or gemcitabine for 6 months. The mFOLFIRINOX regimen consisted of oxaliplatin, leucovorin, and irinotecan administered intravenously at day 1, followed by 5-fluorouracil (5-FU) via contin- uous intravenous (IV) infusion over 46 hours, for 12 two-week cycles. Gemcitabine was administered via IV infusion on days 1, 8, and 15 every 28 days for 6 cycles.

In all, 493 patients were randomized between the 2 regimens. After a median follow-up of 33.6 months, the mFOLFIRINOX regimen demonstrated superiority for OS, disease-free survival (DFS), and cancer-specific survival.

Table. Key Findings in PRODIGE 24 Adjuvant Pancreatic Cancer Trial¹

DFS indicates disease-free survival; mFOLFIRINOX, oxaliplatin, leucovorin, and irinotecan followed by intravenous 5-fluorouracil; MFS, metastasis-free survival; NR, not reached; OS, overall survival.

^aDefined as the first occurrence of any tumor recurrence or metastases, second cancer, or death from any cause.

^bDefined as the time delay between randomization and death due to pancreatic cancer or a treatment-related complication.

The median OS was nearly 20 months longer with mFOLFIRINOX than with gemcitabine: 54.4 versus 35.0 months, which represents a 36% reduction in the risk of death (HR, 0.64; 95% CI, 0.48-0.86; P = .003). The DFS was 8.8 months longer with mFOLFIRINOX than with gemcitabine, 21.6 months versus 12.8 months, respectively. (HR, 0.58; 95% CI, 0.46-0.73; P <.0001). The 3-year DFS was 39.7% with mFOLFIRINOX and 21.4% with gemcitabine. The cancer-specific survival rate was 66.2% for mFOLFIRINOX versus 51.2% with gemcitabine at 3 years (Table).

In terms of safety signals, mFOLFIRINOX resulted in higher rates of grade 3 or 4 adverse events (AEs) than did gemcitabine, respectively, for diarrhea (18.6% vs 3.7%), fatigue (11.0% vs 4.6%), vomiting (5.0% vs 1.2%), and mucositis (2.5% vs 0%) There also was a 9.3% rate of grade 3/4 sensory peripheral neuropathy.

In the gemcitabine arm, the rate of grade 3/4 AEs was higher for thrombocytopenia (4.5% vs 1.3%) and febrile neutropenia (3.7% vs 2.9%). Patients treated with gemcitabine also experienced significantly higher rates of grades 1 to 4 headache, fever, flu-like symptoms, lymphopenia, and transaminase increases. There was 1 toxic death in the gemcitabine arm, said lead author Thierry Conroy, MD, a medical oncologist and director of the Institut de Cancerologie de Lorraine in Nancy, France.

“The gemcitabine arm is better than in every previous trial but the median [overall] survival we got with the FOLFIRINOX regimen is the best ever achieved in any previous adjuvant treatment,” Conroy said. “FOLFIRINOX was more toxic than gemcitabine but it’s still a safe regimen with manageable toxicities. The FOLFIRINOX regimen should be considered the new standard of care after pancreatic cancer resection in patients with good performance status.”

The value of postoperative chemotherapy with gemcitabine and/or fluoropyrimidine for patients with pancreatic cancer has been established for 10 years, Conroy said. Nevertheless, most patients still relapse within 2 years.

In 2011, Conroy and colleagues demonstrated that FOLFIRINOX is more effective than single-agent gemcitabine as a first-line treatment in the metastatic setting for patients with pancreatic cancer and good performance status.⁵ However, Conroy said the toxicities associated with that regimen made it a dif cult choice for adjuvant therapy. Instead, investigators designed the trial with mFOLFIRINOX, which does not include bolus IV 5-FU and is therefore less likely to result in hematologic toxicities and excessive diarrhea.⁶

Although the findings were greeted with enthusiasm, there were several caveats. The population of patients with pancreatic cancer who could potentially benefit from the regimen is relatively small, noted Schilsky. He said only about 15% of patients who present with pancreatic cancer have resectable disease, highlighting the importance of early diagnosis.

Moreover, the population recruited for the trial was “very selected,” Eileen M. O’Reilly, MD, associate director for clinical research at Memorial Sloan Kettering Cancer in New York, New York, said in an interview with OncologyLive^®. “The gemcitabine arm tells us that because this is significantly better than other gemcitabine control arms in other trials,” she said. “Not only were the patients of good performance status, [but] they had to have a CA19-9 less than 180 at the time of study enrollment and they had to be less than 79 years of age.

“Nonetheless, this is probably the biggest signal we’ve ever seen in [pancreatic] cancer that translates into more patients living longer,” she added. “These are exciting data and I would say absolutely practice changing in the clinic for that person who is well enough to tolerate the rigors of that treatment intensity.”

One challenge, Bekaii-Saab said, is that the standard of care in this setting today is gemcitabine plus capecitabine, versus the control arm of gemcitabine alone that was used in PRODIGE 24.

Promising Neoadjuvant Signals

He said investigators are awaiting results from the APACT study, which is testing gemcitabine with and without nab-paclitaxel (Abraxane) as adjuvant therapy in patients with resected pancreatic cancer (NCT01964430). The phase III study, which sought to enroll 866 participants, is active but no longer recruiting participants. The estimated primary completion date is April 2019.In the phase III PREOPANC-1 trial, preoperative treatment with chemotherapy and radiation improved OS rates for patients with resectable or borderline resectable pancreatic cancer compared with immediate surgery. Findings demonstrated that patients treated with neoadjuvant chemoradiotherapy achieved a median OS of 17.1 months compared with 13.7 months with immediate surgery and adjuvant chemotherapy in the intention-to-treat (ITT) population (HR, 0.74; P = .074).

“I must stress that these are preliminary results. We still need 26 more events before the final analysis, according to the statistical plan, can be done, and the final results have to be awaited before we can draw definitive conclusions. But, these preliminary results of PREOPANC suggest a benefit of preoperative chemoradiotherapy over immediate surgery followed by adjuvant chemotherapy,” said lead study author Geertjan Van Tienhoven, MD, PhD, of the Academic Medical Center in Amsterdam, Netherlands.

Currently, the standard of care for patients with resectable or borderline resectable pancreatic adenocarcinoma is surgery followed by adjuvant chemotherapy. Previous studies have suggested the potential for neoadjuvant therapy, yet most of these data were from observational studies, Van Tienhoven commented.

PREOPANC-1 is an ongoing multicenter phase III trial that has randomized 246 patients who were eligible for surgery to either immediate surgery followed by chemotherapy with gemcitabine (n = 127) or neoadjuvant chemotherapy in combination with radiation followed by surgery and chemotherapy (n = 119).

In the investigational arm, patients received 15 treatments of 2.4 Gy radiation combined with 1000 mg/m² of gemcitabine on days 1, 8, and 15, preceded and followed by 1 cycle of gemcitabine. The primary endpoint of the study was OS.

Van Tienhoven stressed that it was important to analyze all patients randomized for treatment in the study (the ITT population) and that findings should not be compared with more specific patient populations. “This is an intention-to-treat analysis of all the patients who have had a diagnosis of resectable pancreatic cancer,” he explained.

As of data cutoff, 142 of the 176 needed events for the primary outcome had been observed. In the immediate surgery arm, 91 patients (72%) received surgery compared with 72 (60%) in the investigational neoadjuvant arm (P = .065). The rate of R0 resection was doubled in the preoperative treatment arm compared with the standard-of-care arm (63% vs 31%, respectively; P <.001).

Van Tienhoven noted that 10% of patients died in the treatment interval between initiation of treatment and surgery, which contributed to the lower rate of resection in the investigational arm. “[The improved R0 resection rate] does indicate that indeed the preoperative treatment does something on the tumor that improves the outcome of surgery,” he said.

Disease progression occurred in 80% of patients who received immediate surgery compared with 50% in the neoadjuvant therapy arm (P = .002). Deaths occurred in 83 patients (65%) in the standard treatment arm compared with 66 (55%) in the neoadjuvant arm (P = .074).

Although Van Tienhoven said that the OS rate was just not statistically significant with a P value of .074, he pointed to the 2-year OS rate of 42% with neoadjuvant therapy, which compared favorably with the 2-year OS rate of 30% from surgery and adjuvant chemotherapy.

Median DFS was 9.9 months with neoadjuvant chemotherapy and radiation therapy versus 7.9 months with surgery and adjuvant therapy (P = .023). The median distant metastasis-free interval was 18.4 months in the investigational arm versus 10.6 months in the standard treatment arm (HR, 0.71; P = .013).

Additionally, the locoregional recurrence-free interval was improved with neoadjuvant therapy compared with surgery and adjuvant therapy (median, not reached vs 11.8 months; HR, 0.55; P = .002). Van Tienhoven noted that differences may stem from the added radiation.

The findings support the concept of neoadjuvant chemotherapy, according to Bekaii-Saab. “Although it still needs to mature a little bit longer, it seems the preoperative approach with chemotherapy and radiation improves outcome. That again makes a strong case for considering neoadjuvant therapy,” he said.

Limitations to the study include the observation that chemoradiotherapy alone is not standard in this setting and that borderline resectable patients should not receive surgery first, Bekaii- Saab said. However, he added, “in the big scheme of things, this tells me that the neoadjuvant approach is a very viable approach and, in my institution, a preferable approach.”

Metastatic Maintenance Therapy

O’Reilly also said the study demonstrates the potential for neoadjuvant chemotherapy, particularly because the patient population is less selective than that of PRODIGE 24. “This is a real-world, more community-based evaluation and it’s on [intention]-to-treat basis, so you’re taking patients at the time of diagnosis as opposed to the very selected patient who’s been able to get through surgery, recover, and is well enough for FOLFIRINOX,” she said. “You can’t compare these 2 studies, but it provides more support for the hypothesis that neoadjuvant therapy in resectable [pancreatic] cancer is an important theme.”In the phase II PRODIGE 35/PANOPTIMOX trial, investigators sought to determine whether an oxaliplatin stop-and-go strategy would be as effective as standard chemotherapy regimens in patients with metastatic pancreatic cancer who have not been previously treated with chemotherapy or radiotherapy. The study was conducted from January 2015 to November 2016 at 53 centers in France.

Participants were randomized to 1 of 3 treatment arms: Arm A, FOLFIRINOX, which contains oxaliplatin, every 2 weeks for a maximum of 12 cycles; Arm B (stop-and-go), FOLFIRINOX during 4 months for 8 cycles, followed by LV5FU2, consisting of leucovorin and 5-FU bolus and infu- sion, as maintenance therapy until progression; and Arm C, sequential treatment alternating FOLFIRI 3, consisting of irinotecan, leucovorin, and 5-FU, with gemcitabine every 2 months. The primary endpoint was the 6-month progres- sion-free survival (PFS) rate.

In all, the study enrolled 273 patients with a median age of approximately 65 years (range, 40-76). Nearly 60% had 1 metastatic site, 28% had 2, and 12% had ≥3. All patients had an ECOG performance score of 0 or 1. Results were reported for the ITT population (n = 273) of all randomized patients and a modified ITT (mITT) cohort (n = 266) of randomized patients who received at least 1 dose of treatment.

The findings demonstrated that the stop-and-go approach employed in Arm B was comparable to the standard FOLFIRINOX regimen used in Arm A. The sequential strategy used in Arm C, which also is called the FIRGEM regimen, was inferior.

In the ITT population, the median PFS, defined as first progression from any chemotherapy received, was 6.3 months (95% CI, 5.3-7.6) in Arm A (n = 91); 5.7 months (95% CI, 5.3-7.5) in Arm B (n = 92); and 4.5 months (95% CI, 3.5-5.7) in Arm C (n = 90).

The median OS was 10.1 months (95% CI, 8.5-12.2) in Arm A, 11.0 months (95% CI, 8.7-13.1) in Arm B, and 7.3 months (95% CI, 5.7-9.5) in Arm C. The 18-month OS rates were 18.5%, 28%, and 13.9% for the 3 arms, respectively, yielding a P value >.05 in favor of the stop-and-go strategy in an exploratory analysis.

In the mITT population, the 6-month PFS rate was 47.1% in Arm A (n = 87), 44.0% in Arm B (n = 91), and 34.1% in Arm C (n = 88).

The most common grade 3/4 AEs for Arms A, B, and C, respectively, included neutropenia at 28.4%, 25.3%, and 32.2%; asthenia at 25.0%, 30.8%, and 32.2%; and diarrhea at 11.4%, 17.6%, and 18.4%. Of note, grade 3/4 sensory neuropathy was markedly higher with stop-and-go therapy in Arm B at 18.7% compared with 10.2% in Arm A and 0% in Arm C.

Laetitia Dahan, MD, of La Timone University Hospital in Marseille, France, and colleagues concluded that FOLFIRINOX with LV5FU2 as a maintenance strategy is a “feasible and effective” regimen. They noted that the rate of severe neurotoxicity was higher than expected in the maintenance arm because of a higher cumulative dose of oxaliplatin, and that grade 3/4 neurotoxicity was observed later in the course of treatment.

“Interestingly, [the results of] this study showed that the benefit was very similar in both arms,” said Bekaii-Saab. “What that tells us, essentially, is that we don’t have to expose our patients to FOLFIRINOX until those toxicities hit or until progression. We can limit to 3 months and then shift to 5-FU.”

The decision of whether to repeat treatment with an oxaliplatin regimen could be made based on the risk of neuropathy, Bekaii-Saab added. He noted that the phase III POLO study is exploring the efficacy of maintenance olaparib (Lynparza) in patients with germline BRCA-mutated pancreatic cancer whose disease has not progressed after first-line platinum-based chemotherapy (NCT02184195). “This question of maintenance has become very relevant in our world,” he said.

The PRODIGE 35/PANOPTIMOX results, O’Reilly said, suggest a way to de-escalate the intensity of the chemotherapy by dropping oxaliplatin and then reintroducing it at the time of progression.

“That’s a theme that resonates with what we know in colorectal cancer and how we hope to be increasingly able to treat patients with [pancreatic] cancer,” she said. “These therapies are tough and they’re cumulatively debilitating for patients in terms of fatigue, myelosuppression, and neuropathy. To be able to deintensify therapy and offer patients—not a break in the full sense of the word but maybe an improvement in the quality of life for a period of time—hopefully, for a subset of people, will have a value.”

References

1. Conroy T, Hammel P, Hebbar M, et al; CCTG and the UNICANCER-GI/PRODIGE Group. Unicancer GI PRODIGE 24/CCTG PA.6 trial: a multicenter international randomized phase III trial of adjuvant mFOLFIRINOX versus gemcitabine (gem) in patients with resected pancreatic ductal adenocarcinomas. Presented at: 2018 ASCO Annual Meeting; June 1-5, 2018; Chicago, IL. Abstract LBA4001. meetinglibrary.asco.org/record/159164/abstract.e
2. Van Tienhoven G, Versteijne E, Suker M, et al; Dutch Pancreatic Cancer Group. Preoperative chemoradiotherapy versus immediate surgery for resectable and borderline resectable pancreatic cancer (PREOPANC-1): a randomized, controlled, multicenter phase III trial. Presented at: 2018 ASCO Annual Meeting; June 1-5, 2018; Chicago, IL. Abstract LBA4002. meetinglibrary.asco.org/record/160063/abstract.
3. Dahan L, Phelip JM, Le Malicot K, et al. FOLFIRINOX until progression, FOLFIRINOX with maintenance treatment, or sequential treatment with gemcitabine and FOLFIRI.3 for first-line treatment of metastatic pancreatic cancer: a randomized phase II trial (PRODIGE 35-PANOPTIMOX). Presented at: 2018 ASCO Annual Meeting; June 1-5, 2018; Chicago, IL. Abstract 4000. meetinglibrary.asco.org/record/159166/abstract.
4. Cancer stat facts: pancreatic cancer. National Cancer Institute website. seer.cancer.gov/statfacts/html/pancreas.html. Updated April 2018. Accessed July 5, 2018.
5. Conroy T, Desseigne F, Ychou M, et al; Groupe Tumeurs Digestives of Unicancer and PRODIGE Intergroup. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011;364(19):1817-1825. doi: 10.1056/NEJMoa1011923.
6. Mahaseth H, Brutcher E, Kauh J, et al. Modified FOLFIRINOX regimen with improved safety and maintained efficacy in pancreatic adenocarcinoma. Pancreas. 2013;42(8):1311-1315. doi: 10.1097/MPA.0b013e31829e2006.