Locally Advanced and Metastatic Pancreatic Cancer: Survival analysis and prognostic factors in a case series

Pancreatic cancer represents the seventh leading cause of cancer mortality in the world, and the second leading cause of cancer death in the United States1. Most patients with such neoplasm present at an advanced stage of the disease at diagnosis, approximately 50% with metastatic disease, and 25% to 35% with borderline or locally advanced disease2. Gemcitabine plus nab-paclitaxel or FOLFIRINOX represent the first line of chemotherapy treatment in locally advanced and metastatic disease with median overall surival of 8.5 and 11.1 months, respectively3 4. A limitation to be considered concerning randomized controlled trials is the fact that their eligibility criteria are stringent and, as a result, only a small portion of patients with advanced pancreatic cancer is considered able to participate in the studies. Not surprisingly, such a highly select study population generally does not accurately reflect the population of patients in need of treatment5.

Due to the low survival associated with this neoplasia, different research has directed efforts to identify predictive markers that help to optimize the treatment. Among them, the carbohydrate antigen 19-9 (CA19-9), the performance index evaluated by the ECOG (Eastern Cooperative Oncology Group), age, local therapies, and the status of chronic inflammation represented by the neutrophil/lymphocyte (NLR) and platelet/lymphocyte (PLR) ratio6 7.

These last two have been particularly studied due to the advance in the knowledge of immunology in cancer. The NLR is calculated from the absolute values of neutrophils and lymphocytes obtained through the blood count. Many studies indicate that a low NLR (usually less than four or five) is associated with better survival in patients with metastatic pancreatic cancer8 9. PLR (platelet/lymphocyte ratio), calculated from the absolute values of platelets and lymphocytes, was also an unfavorable predictor of overall survival and progression-free survival in patients with pancreatic cancer10.

Finally, has been studied if local therapy can improve survival in patients with locally advanced pancreatic cancer. A series of cases from a retrospective study by the Johns Hopkins group, for example, showed that patients with locally advanced pancreatic cancer who underwent radiation therapy, at some point in their treatment, had a median overall survival of 13.7 months and 61% disease control in one year11 and two other studies that used radiotherapy in patients with locally advanced disease reached a median overall survival of 15 months and 12.2 months, respectively12 13. Nevertheless, a large randomized phase III study LAP07 did not show any benefit for radiotherapy being added to chemotherapy4.

The present study evaluated the clinical and epidemiological characteristics and the factors that influenced the survival of patients with metastatic and unresectable locally advanced pancreatic adenocarcinoma, treated in a private reference hospital for the treatment of cancer in the Brazilian state of Pernambuco.

The research was approved by the Human Research Ethics Committee of Pernambuco's Real Hospital Português de Beneficência. Opinion Number: 3.705.999.

The present study observed a median overall survival of 13.93 months in patients with unresectable and metastatic locally advanced pancreatic adenocarcinoma-results superior to those found in the two main studies in this group of patients. Those studies observed a higher overall survival for FOLFIRINOX (11.1 months) and the combination gemcitabine plus nab-paclitaxel (8.5 months) in the first line of treatment when compared with gemcitabine alone3 4, while our sample had a median survival of 12 months and 13 months, respectively. A study conducted at Instituto do Câncer do Estado de São Paulo, involving 61 patients with advanced pancreatic adenocarcinoma, 31 of whom with metastatic disease, had similar survival to that found in our series15.

The ability to predict poor survival for patients with a disease as severe as pancreatic cancer is essential in the initial management before treatment to select patients most likely to receive polychemotherapy such as FOLFIRINOX and gemcitabine plus nab-paclitaxel. For that, possible factors associated with worse survival were evaluated, such as age, disease stage, ECOG, NLR, PLR, CA 19-9, and the use of local therapy.

The median age observed was 68 years old, above that reported in the PRODIGE and MPACT studies, which were 61 and 62 years old, respectively3 4. In the present study, we did not note a difference in overall survival when considering the age of 65 years (p=0.27) or 75 years (p=0.16). It's known that in older patients with numerous comorbidities, the treatment of advanced malignancy, such as pancreatic cancer, may not result in clinical benefit and increase the risk of toxicities, clinical decompensation and undertreatment, leading to worse overall survival16. Despite this, age has not been a predictor of worse survival in patients selected for clinical trials7 17.

Our sample did not show any difference in overall survival when comparing the locally advanced and metastatic clinical stages (13.9 months x 14.8 months; p=0.653). However, despite data in the literature pointing to better survival in locally advanced disease17 18 19, such survival is similar to that observed in our patients.

About one-third of our population had ECOG 2 (n=11, 31.4%), which is different from that observed in clinical studies, where the majority of patients are recruited with better performance. Therefore, more real-life studies are essential to assess the impact of treatments on clinical practice. In the MPACT study 8% of the patients included had ECOG 2 and this group of patients had a worse survival when compared with those with better ECOG6 7. Our sample showed a median overall survival of 3.9 months that is similar to that observed in the MPACT study for the group of ECOG 2 patients who received gemcitabine plus nabpaclitaxel. Likewise, patients with better performance status also had better overall survival (20.4 months x 3.9 months; p=0.021).

Another prognostic factor that has been widely studied is the inflammatory factor. The search for markers of systemic inflammation measured by pretreatment NLR and PLR is a prognostic factor in several studies, which point to a worse survival outcome in patients with elevated NLR and PLR8 9 10. In our series, a cutoff point indicated by the literature was used and it was observed that patients with an NLR of less than four survived almost twice as long when compared with those with an NLR greater than or equal to four (15.6 months x 8.5 months; p=0.029). However, concerning PLR, when using the cut-off point of 150, as suggested in the literature10, no difference in overall survival was shown.

The CA 19-9 marker is also studied as a prognostic factor for pancreatic cancer, with contradictory results and variable cut-off points. In some studies, a high value before treatment predicts advanced disease, unresectability, recurrence, and worse overall survival20 21 22 23. On the other hand, Tabernero et al. studied this marker as a prognostic factor in patients receiving nab-paclitaxel and gemcitabine and found no correlation with survival7. In the present study, the median value of CA 19-9 was 256.98 U/ml. Values above or below the standard reference (37 U/ ml, data not shown) or close to the median, 250 U/ ml, were analyzed as cut-off points and were not related to overall survival (p=0.262 and p=0.519, respectively).

With regard to local therapy, some studies indicate better survival in patients with advanced disease who receive local treatment (radiotherapy or radioablation) at some point in their treatment11 12 13 24. Local treatment was used in seven patients in our study, radiotherapy (n=6) and radioablation (n=1), during the treatment of locally advanced (n=4) or metastatic (n=3) disease. These patients survived significantly longer when compared to those who did not (p=0.006). Despite our results, a large randomized trial LAP07, did not shows any benefit for the use of radiation in this setting14.

As limitations for the present study, we can mention the retrospective and single- centered design and the small number of participants. On the other hand, our study presents information from our daily clinical practice with survival data similar to those observed in phase III clinical studies.

In conclusion, good clinical condition (ECOG 0 or 1), low NLR, and local therapy are associated with better survival in univariate analysis, suggesting their clinical usefulness and applicability in the treatment of locally advanced or metastatic pancreatic cancer.