The present study focused on the construction of a nomogram for prognosis prediction in metastatic or locally advanced unresectable PC patients treated with first-line systemic chemotherapy. The clinical variables included liver metastases, local therapy, operation, L3SMI change per 2 months and CD4:CD8 ratio for PFS and CA19-9, local therapy, operation, weight change per 2 months and CD4:CD8 ratio for OS. Here, for the first time, we demonstrate that dynamic changes in nutritional status and the proportion of immune cells can be used for prognosis prediction. In the setting of an advanced state with limited treatment options, screening for patients who may benefit from chemotherapy is important.
Pancreatic cancer shows the second-highest malnutrition rate (47.6%), and early nutritional intervention is indispensable in PC patients receiving chemotherapy.(16) ONS appeared to be effective and convenient in improving nutritional status by maintaining body composition and alleviating fatigue symptoms.(16) Irene et al. found poor adherence to ESPEN (European Society Of Parenteral Enteral Nutrition) guidelines in the real world.(17) In our research, the proportion of patients receiving ONS was lower than that of patients at nutritional risk. Almost half of the patients received ONS, and their weight and skeletal muscle loss were greatly improved compared to the non-ONS group. Since non-ONS was associated with negative outcomes in our study, adherence to evidence-based nutritional recommendations is highly desirable. Although cancer cachexia cannot be treated with nutritional therapy alone, optimal nutritional care is recommended as a cornerstone of multimodal therapy.(18)
There seems to be no consensus among different studies on how weight change affects the prognosis of PC.(19–21) The conclusion of the effect of skeletal muscle on the prognosis of PC seems to be more consistent.(22–25) In our study, skeletal muscle change was identified as a prognostic factor in PFS nomograms by LASSO Cox regression analysis, whereas weight change only showed borderline significance in the final model of OS. There is a clear shift in the definition of cancer cachexia that requires oncologists to assess muscle loss, rather than simply weight loss, which may be masked by fat gain, additional water in the form of oedema, ascites, or pleural effusion.(26) It has been well reported that chemotherapy affects body composition throughout the treatment course and was estimated to result in a relative muscle loss of 2.9% every 100 days for PC patients undergoing chemotherapy.(27) Skeletal muscle loss also increases the toxicity of chemotherapy, which, in turn, leads to chemotherapy delay or reduction.
In our study, the peripheral CD4+:CD8 + T lymphocyte ratio at baseline was associated with PFS and OS. Inflammation is closely correlated with immunity, and immune cell populations contribute to both inflammation and the immune response, which enhance tumour growth and metastasis.(28) Wang et al. reported that CD4 + and CD8 + T cells in the microenvironment of pancreatic cancer were the principle risk factors for prognosis.(29) The involvement of both CD4 + and CD8 + T cells is required for an effective antitumour immune response.(30) IMpassion130, a phase III multicentre, randomized, double-blind, placebo-controlled trial in metastatic triple-negative breast cancer patients, showed that high CD8 + TIL expression was significantly associated with improvements in OS and PFS.(31) In our study, we focused on patients at advanced stages. Therefore, it was difficult to obtain tumour tissues to discover the local immune status. The peripheral lymphocytes could reflect locally infiltrating lymphocytes to some extent.
The liver, as the most common metastatic site for PC, was correlated with poor prognosis.(32) When the tumour grows extremely in the liver, life-threatening symptoms such as jaundice, abnormal liver function, ascites, liver failure, and hepatic encephalopathy often occur. Almost all patients with a survival period of less than 4 months (11 patients) had liver metastasis in our study. Although the first-line systemic chemotherapy regimen in this study cannot be unified, we obtained a disease control rate of 78.9% and an overall response rate of 15.6%. Based on the results of multidisciplinary discussions, the appropriate patient can receive local therapy such as radiotherapy or radiofrequency ablation. Among 11 patients who received successful conversion therapy and surgical resection, two patients remain in the stage of no evidence of disease until now. Patients who have the opportunity to undergo local treatment or surgical resection often have a relatively small tumour burden and are in good physical condition. At the same time, through the above method, we can achieve the purpose of reducing the tumour to a great extent. Accordingly, our multivariate analysis indicated that local therapy and surgery were also powerful in predicting patient survival. A multidisciplinary comprehensive treatment model has greatly improved the prognosis of PC patients.
Several clinical variables have been reported to predict prognosis in metastatic or locally advanced unresectable PC, including performance status(12, 14, 33), tumour size or stage(12, 13, 33, 34), and serum biomarkers such as CA19-9(12–14, 33), albumin(13, 35, 36), and neutrophil-to-lymphocyte ratio(14, 33, 35)4. CA19-9 is widely recognized as a prognostic factor with good sensitivity and specificity for PC, and this was also confirmed in our study. However, some patients cannot secrete CA19-9 into their serum with negative red blood cell phenotypes on both Lewis A and B antigens. Because most of the patients in this study were in stage IV, tumour staging was not included in the prognostic analysis. Compared with previous PC nomograms, we first incorporated dynamic changes in nutritional status during chemotherapy into the analysis.
We are aware of several limitations in our study. First, its retrospective study design and small sample size limit the generalization of the results. Although 100 bootstrap resamplings were performed to validate this model, external validation of cohorts from other centres was unavailable in our study. Second, due to incomplete data, the amount of ONS intake was not classified. Therefore, multicentre prospective randomized clinical trials with large sample sizes are needed to confirm our results.
In conclusion, nomograms were developed from clinical features and nutritional indicators to predict the probability of achieving 2-month and 4-month PFS and 6-month and 12-month OS with chemotherapy for advanced PC. These nomograms may be useful to better predict expected survival time.