Metastatic pancreatic cancer with complete response to FOLFIRINOX treatment

Abstract

A 59-year-old woman presented with abdominal pain associated with nausea and night sweats. A large mass was found in the pancreatic tail and innumerable liver lesions were identified. Ultrasound-guided biopsy of a liver nodule confirmed moderately differentiated adenocarcinoma consistent with a pancreatobiliary primary. On FOLFIRINOX chemotherapy, subsequent CT scans showed shrinkage of the pancreatic mass and liver metastases. Her cancer antigen 19-9 (CA 19-9) normalised after 11 months. Oxaliplatin was discontinued due to peripheral neuropathy but she completed 37 cycles of FOLFIRI during which her pancreatic mass disappeared, liver lesions decreased in size and were subsequently deemed to be scar tissue by the radiologist. After 4 years of treatment, the patient agreed to a break from chemotherapy. Eighteen months afterwards, an MRI abdomen continues to demonstrate no visible pancreatic mass and the two remaining liver lesions, believed to be scar tissue, remain stable. Her CA 19-9 level remains normal. This appears to be a complete response to FOLFIRINOX/FOLFIRI chemotherapy in a patient with metastatic pancreatic cancer.

Background

Metastatic pancreatic cancer traditionally has a very poor prognosis. Of all patients diagnosed with pancreatic cancer, only 50% survive past 4 months, and the 5-year survival rate for patients with metastatic disease is only 8% in Canada and 7.3% in the UK.^{1 2} The majority of pancreatic cancers are diagnosed in an advanced stage when surgery is not possible. In addition, most pancreatic cancers are highly unresponsive to chemotherapy and radiation.³

For more than a decade, single-agent gemcitabine was the standard of care for systemic treatment of advanced pancreatic cancer. This was based on a pivotal randomised trial which showed that gemcitabine treatment resulted in a response rate of 5%, clinical benefit response of 24% and 1-year survival of 18%.⁴ Fourteen years later, the ACCORD 11 trial established a new option for first-line treatment of metastatic pancreatic cancer with the FOLIFIRINOX (oxaliplatin, irinotecan, fluorouracil (5-FU) and leucovorin) regimen. Compared with gemcitabine monotherapy, it showed a superior response rate (31% vs 9%) and progression-free survival (6.4 vs 3.3 months).⁵ Despite the higher response rate, median overall survival (OS) with FOLFIRINOX therapy was still only 11 months.⁵ Complete responses (CRs) were rare, with only one reported in this pivotal phase III trial, but details were not provided.⁴

In this report, we present the case of a 59-year-old woman who presented with biopsy-proven metastatic pancreatic adenocarcinoma with liver metastases. After treatment with FOLFIRINOX, and subsequently FOLFIRI, she achieved a radiographic CR and is physically well more than 5 years from her initial diagnosis of metastatic pancreatic cancer.

Case presentation

A previously healthy 59-year-old woman presented with a 2-month history of right upper quadrant abdominal pain and cramping associated with nausea. Over a period of 3 weeks, her abdominal pain became progressively worse: it began to radiate to the epigastric area, became more frequent and intense as well as being associated with drenching night sweats. This pain was also associated with significant weight loss of 7 pounds in 2 weeks. Her medical history is unremarkable, consisting of only migraine headaches. Her family history was significant for one relative with prostate cancer and another with breast cancer. She previously smoked 10 cigarettes per day for 25 years.

At the time of her medical oncology consultation, her Eastern Cooperative Oncology Group (ECOG) performance status was 1. On examination, her abdomen was soft and mildly tender in the right upper quadrant on palpation. She also had a tender and warm right calf. The rest of her physical examination was unremarkable.

Investigations

The patient was sent for an abdominal ultrasound by her family doctor, and was found to have multiple hypoechoic liver lesions, ranging from 1.8 to 3.2 cm in size. She was then sent for a contrast-enhanced CT scan of the chest, abdomen and pelvis. The scan showed a large hypoattenuating mass in the pancreatic tail (3.2×2.8 cm), consistent with a primary pancreatic adenocarcinoma. There were more than 30 liver lesions, ranging from 1 to 6 cm in size (figure 1). The CT scan also showed bilateral pulmonary thromboembolism for which the patient was admitted to hospital and started on anticoagulation with low molecular weight heparin.

Figure 1.

Portal venous phase contrast-enhanced CTs. Initial CT (January 2015) demonstrates mass in the pancreatic tail and multifocal liver metastases (left, top and bottom). Post-treatment CT (December 2019) demonstrates complete resolution of pancreatic tail mass and liver metastases (right, top and bottom).

While in the hospital, her blood work revealed a markedly elevated cancer antigen 19-9 (CA 19-9; 14 514 kU/L; reference range <=35 kU/L; figure 2). An ultrasound-guided biopsy of one of the large liver lesions was obtained, which confirmed a moderately differentiated adenocarcinoma consistent with a pancreaticobiliary or upper gastrointestinal tract primary cancer (figure 3).

Figure 2.

Cancer antigen 19-9 (CA 19-9) levels in our patient over the course of treatment with FOLFIRINOX/FOLFIRI (reference range <=35 kU/L). CA 19-9 was markedly elevated at diagnosis (14 514 kU/L), peaked at 35 170 kU/L and normalised there afterwards, remaining so throughout the rest of her treatment.

Figure 3.

Core biopsy of liver tumour with adenocarcinoma showing small glands (long arrow), small mucin collections (star), moderate nuclear pleomorphism and mitoses (short arrow).

Treatment

The patient was seen by a medical oncologist and informed that her metastatic pancreatic cancer was incurable. It was discussed that the goal of palliative chemotherapy was to prolong survival and hopefully maintain a reasonable quality of life. The decision was made to start FOLFIRINOX chemotherapy with palliative intent given the potential for inducing a response and prolonging survival in metastatic pancreatic cancer. The patient was started on FOLFIRINOX at 50% of the recommended dose to test tolerance, and slowly increased to 100% by the fourth cycle. She was also prescribed pegfilgrastrim as primary prophylaxis for febrile neutropenia. Her CA 19-9 peaked at 35 170 kU/L prior to her second cycle of FOLFIRINOX and then began trending downwards (figure 2). After six cycles of treatment over 3 months, her CT showed a decrease in the size of her pancreatic mass and the large liver lesions, as well as resolution of the smaller liver lesions. The patient experienced some nausea, loss of appetite and fatigue, but these symptoms were manageable. She experienced some abdominal pain throughout her treatments, which was controlled with morphine. She tolerated FOLFIRINOX relatively well and completed a total of nine cycles.

After her ninth cycle of FOLFIRINOX, she developed worsening fatigue and neuropathy in her fingertips. Oxaliplatin was discontinued, but the patient continued treatment with FOLFIRI. She tolerated FOLFIRI well, experiencing only fatigue at certain points throughout her treatment. The cancer continued to shrink on subsequent CT scans and after 1 year and 9 months of chemotherapy, the primary pancreatic cancer mass could no longer be seen on CT images. Her CA 19-9 continued to trend down and normalised after eleven months ofFOLFIRINOX/FOLFIRI treatment (figure 2). The patient continued on FOLFIRI for a total of 37 cycles, and her liver metastases continued to decrease in size. By December 2019, there was complete resolution of the pancreatic mass and disappearance of all but two liver lesions on CT (figure 1).

Given the persistence of two small liver lesions on CT, her case was discussed at a multidisciplinary hepatobiliary tumour board to determine whether localised treatment of the lesions should be considered. Based on the CT and contrast-enhanced ultrasound, the tumour board radiologists were of the opinion that these two liver lesions represent residual scar tissue and no localised treatment was warranted.

When informed that the multidisciplinary tumour board thought her metastatic pancreatic cancer has achieved a radiographic CR, the patient decided to take a break from FOLFIRI chemotherapy. A CT scan was repeated at 3, 6 and 12 months after stopping chemotherapy and each demonstrated no evidence of progressive metastatic disease. The pancreatic mass remains undetectable, and the two liver lesions are stable or slightly decreased in size. An MRI of her abdomen 18 months after cessation of chemotherapy confirms treated liver metastases with no definite residual disease. Her CA 19-9 has remained normal at 26 kU/L(reference range <=35 kU/L). Clinically, she remains completely well.

Outcome and follow-up

The patient is currently alive and well more than 5 years after her initial diagnosis of metastatic pancreatic cancer and maintains a radiographic CR. She has been counselled to watch for recurrence of clinical symptoms. She remains clinically well and will be followed up with another CT scan in 6 months. Her CA 19-9 will be monitored every 3 months.

Considering this remarkable response, we performed a FoundationOne CDX test on her tumour tissue to better evaluate the tumour biology. The results showed that her cancer is microsatellite-stable, has a Cyclin D1 (CCND1) amplification, a mutation in Kirsten rat sarcoma virus oncogene (KRAS; G12D) and tumor protein p53 (TP53; G245S). No breast cancer gene (BRCA) or partner and localizer of BRCA2 (PALB2) mutations were found.

Discussion

The above describes a rare case of metastatic pancreatic cancer with a CR while receiving FOLFIRINOX/FOLFIRI chemotherapy. Metastatic pancreatic adenocarcinoma is usually a fatal and aggressive disease. CR of metastatic pancreatic cancer is rare, although five other reports of patients with an outstanding response to FOLFIRINOX/FOLFIRI chemotherapy alone have also been published.^6–10

In a case reported by Tsujie and associates, a patient with pancreatic cancer and distant metastases to the para-aortic lymph nodes and Virchow’s node was found to have pathological CR on assessment of tissue from conversion surgery after treatment with FOLFIRINOX/FOLFIRI.⁶ Like our case, the tumour did not have a BRCA1/2 mutation, and the positive response to FOLFIRINOX chemotherapy was felt not to be related to BRCA1/2 mutations.⁶ There was a second report showing CR in a patient with metastatic pancreatic adenocarcinoma.⁷ Similar to our case, the patient was treated with FOLFIRINOX and stepped down to FOLFIRI due to peripheral neuropathy, but unfortunately, genetic testing was not performed and we are unable to compare genetic profiles of the cases.⁷ Luu et al published a case of unresectable pancreatic cancer with liver metastases treated with FOLFIRINOX therapy with palliative intent.⁸ This also resulted in a CR, but the authors were unable to pinpoint a causative factor for the response.⁸ According to Schneitler et al, two patients with metastatic pancreatic cancer achieved CR with FOLFIRINOX but when maintained on different regimens (patient 1 on gemcitabine and patient 2 on FOLFIRINOX), only the patient who received gemcitabine experienced recurrence.⁹ Lastly, Chung et al, evaluated the efficacy of FOLFIRINOX as a second-line agent in pancreatic cancer that failed to respond to gemcitabine.¹⁰ One patient with metastatic pancreatic cancer achieved a CR, and FOLFIRINOX was effective as an alternative regimen for gemcitabine-resistant pancreatic cancers.¹⁰

In another report, Ozaki et al described a patient with pancreatic ductal adenocarcinoma and liver metastases who was treated with FOLFIRINOX and downstaged to the point where curative surgery was possible.¹¹ If FOLFIRINOX had been continued, perhaps a CR would have been achieved since the there was a pathological CR of the liver metastases in this case.¹¹

Genomic testing and gene profiling are important initial steps in the management of exocrine pancreatic cancer. Pancreatic cancer is the third most common type of cancer associated with BRCA1/2 mutations, and mutations in genes involved in DNA damage repair like BRCA1/2 and PALB2 are often associated with better response rates, with increased sensitivity to platinum-based regimens and poly(ADP-ribose) polymerase inhibitors.^12–14 Patients with these mutations were found to have a better OS (16.8 vs 9.1 months, p=0.03) and overall response rate (58% vs 21%, p=0.0022).^{12 15} Surprisingly, there were no BRCA1/2 and PALB2 mutations and no other known genetic alteration present on the FoundationOne CDX test that could explain the exceptional response achieved in our patient’s case.

KRAS mutations are present in 75%–100% of pancreatic cancers, and the G12D KRAS mutation is the most common type seen in pancreatic cancer.¹⁶ This mutation confers no particular sensitivity to FOLFIRINOX/FOLFIRI chemotherapy, as the type of KRAS mutation does not usually have a differential effect on cancer growth and dissemination.¹⁶ The TP53 gene is often inactivated in pancreatic cancer, as seen in our patient’s tumour.¹⁷ TP53 positive cancer often presents at a more advanced clinical stage, has a worse postoperative prognosis and can result in 5-FU resistance.¹⁷ Similarly, CCND1 amplifications were shown to confer chemoresistance and promote cellular proliferation, and have been associated with poor prognosis in pancreatic cancer.^18–20 In our patient’s case, we could not identify any genetic cause for the excellent response observed.

As expected, significant neuropathy eventually developed due to oxaliplatin and it was discontinued. The patient continued treatment with FOLFIRI and we observed a continued response radiographically and biochemically with decreasing CA 19-9 levels. The efficacy of oxaliplatin in metastatic pancreatic cancer has been questioned and there is evidence of efficacy with irinotecan-based regimens in the absence of oxaliplatin.^{21 22} In another case report, the authors also questioned whether their patient’s CR was due to induction with FOLFIRINOX or if there was a significant effect from FOLFIRI on the cancer cells.⁷

There are no standardised follow-up guidelines since CRs are exceedingly rare in metastatic pancreatic cancer. Our patient is doing well clinically and therefore we have opted to monitor her with repeat CT scans every 6 months and CA 19-9 levels every 3 months.

In conclusion, we present an exceptionally rare case of metastatic pancreatic adenocarcinoma which sustained a CR after FOLFIRINOX/FOLFIRI treatment. This was despite the absence of known genetic characteristics which would predict a better response to treatment or improved prognosis. Further studies are needed to better understand the complex biology of exocrine pancreatic cancer and to identify molecular abnormalities that can act as predictive and prognostic factors for patients with this disease.

Learning points.

• Complete response (CR) in metastatic pancreatic cancer is rare, but possible.

• CR in metastatic pancreatic cancer is not always associated with ‘susceptible’ mutations.

• Further investigation and follow-up of patients with exceptionally positive responses are needed to improve understanding of metastatic pancreatic cancer and make positive outcomes more common.