Albumin-to-alkaline phosphatase ratio serves as a prognostic indicator in unresectable pancreatic ductal adenocarcinoma: a propensity score matching analysis

Pancreatic ductal adenocarcinoma (PDAC) is a fatal disease and the fourth leading cause of cancer-related death worldwide with a 5-year survival rate of 8% [1]. In China, PDAC is ranked sixth among cancer-related deaths and ninth among the most prevalent cancers [2]. Currently, there are no effective treatments for PDAC. Surgical resection remains the only curative treatment. However, the overall management of PDAC patients is far from being satisfactory, since approximately 80% of patients are diagnosed at advanced and unresectable stages and received systemic chemotherapy [3, 4]. Gemcitabine-based combination regimen or the fluorouracil, leucovorin, irinotecan, and oxaliplatin (FOLFIRINOX) are currently the most recommended in cases of advanced PDAC [5‐7]. Unfortunately, tumor biology of PDAC causes resistance to systemic therapy, thereby leading to poor prognosis. Notably, patients with similar PDAC disease conditions sometimes have different outcomes. Therefore, it is crucial to identify a prognostic marker that can predict survival and hence guide clinical decisions for patients with unresectable PDAC.

Serum albumin (ALB) is synthesized in the liver and is the most abundant protein in plasma and this may explain why it is usually included in standard blood tests to assess hepatic function. Also, it performs various roles, ranging from regulating oncotic pressure to scavenging reactive oxygen species. Recent evidence has shown that ALB is an accurate biomarker reflecting underlying host systemic inflammatory responses [8]. Beside that, alkaline phosphatase (ALP) is enriched in kidney, bile duct, and liver [9]. Studies have shown that serum ALP is a useful marker of tumor growth and metastasis [10‐12].

Accordingly, albumin-to-alkaline phosphatase ratio (AAPR) is postulated to be a prognostic indicator in hepatocellular carcinoma [13]. So far, its prognostic performance has been studied for multiple malignancies such as non-small-cell lung cancer [14], metastatic nasopharyngeal cancer [15], cholangiocarcinoma [16], upper tract urothelial carcinoma [17], renal cell carcinoma [18], and cervical cancer [19]. Notably, Pu et al. [20] concluded that preoperative AAPR is an independent factor that determines PDAC progression after curative resection. However, little is known about the ability of AAPR to reflect the clinical outcomes of unresectable PDAC, which accounts for the majority of the PDAC.

Taking this need into consideration, we conducted a retrospective cohort study and performed propensity score-matching (PSM) analysis to investigate the prognostic significance of pretreatment AAPR in unresectable PDAC treated with chemotherapy.

In the current study, we constructed AAPR, an easily available index comprised of serum ALB and ALP level. We further investigated its prognostic significance in unresectable PDAC patients. Analysis of the patient characteristics showed that low AAPR was closely associated with more advanced stage and higher CA19–9 level. PSM analysis was performed to generate well-balanced patients in low and high AAPR groups for survival comparison. Univariate and multivariate Cox analysis indicated that APPR served as an independent prognostic predictor for OS in both entire cohort and PSM cohort. In addition, we found that AAPR could be used in most subgroups except for the group of age ≤ 60. More importantly, inclusion of AAPR improved the prediction performance of conventional prognostic model.

To our best knowledge, Chan et al. [13] were the first to report that AAPR can act as a novel prognostic index for patients with hepatocellular carcinoma patients underwent curative surgery. Since then, numerous studies have looked into the prognostic performance of AAPR for various malignancies, such as metastatic nasopharyngeal cancer [15], non-small-cell lung cancer [14], cholangiocarcinoma [16], upper tract urothelial carcinoma [17], renal cell carcinoma [18], and cervical cancer [19]. Particularly, Pu et al. [20] examined the AAPR in patients receiving surgery for PDAC and concluded that preoperative AAPR could independently predict postoperative OS in PDAC patients. Consistently, our data revealed that low AAPR correlated with poor OS. Also, we revealed that AAPR might function as an independent prognostic predictor in unresectable PDAC patients. In the subgroup analyses, the outcome of survival consistently favored the high AAPR group across most subgroups except in the group of age ≤ 60. Despite the findings not being significant, the trend in the group of age ≤ 60 was consistent with other groups. In addition, it was reported that patients with low AAPR had less favorable pathological features, including higher rations of lymph node metastasis and advanced T stage in non-small-cell lung cancer [14]. Low AAPR was reported to be associated with larger tumor size and more lymph node positive status in upper tract urothelial carcinoma [17]. Similarly, we found that patients with low APPR levels were more likely to have metastatic disease and higher CA19–9 level. These findings suggested that low AAPR was associated with progress of disease, which called for attention from clinicians.

Cancer is marked by rapid tumor cell proliferation and invasion [30], which may lead to various metabolic changes, such as enhanced production of some serum proteins, cytokines, and hormones. Thus, serum ALB and ALP might reflect tumor progression, and can, therefore, be used to predict clinical outcomes. The following mechanisms may explain the biological relationship between AAPR and prognosis of unresectable PDAC.

As the most abundant protein in serum, ALB is the primary determinant of plasma oncotic pressure. Besides, ALB offers storages and conveyors for many exogenous and endogenous substances [31]. Serum ALB levels reflect the nutritional status and liver function in humans. It is widely recognized that inflammation, tumor microenvironment, and their interaction play critical roles in tumor progression [32]. Studies have shown that ALB can modulate the inflammatory reaction by binding prostaglandins, lipopolysaccharide, and reactive oxygen species [33]. Moreover, ALB has been found to suppress the cell cycle and progression of hepatocellular carcinoma [34, 35]. Recent lines of evidence have identified ALB as an independent prognostic indicator in various malignant tumors, including lung [36], breast [37], and gastric cancers [38].

As a phosphate monoester hydrolase, ALP promotes the hydrolysis and transfer of phosphate groups in alkaline conditions [39]; therefore, it is regularly examined to evaluate liver function in clinical practice. Increasing evidence has shown that ALP can serve as a tumor-associated antigen and biomarker of cancer cell proliferation [11, 40]. On the contrary, a reduction of ALP activity induces cell death, mesenchymal-to-epithelial transition, and suppresses cell migration [41]. Besides, ALP activity level is commonly used to predict bone metastasis in various types of cancer [42, 43]. In the present study, patients with lower AAPR displayed poor prognosis, an observation that was consistent with the previous findings.

As a result of inflammation, oxidative stress produces reactive oxygen species that can damage protein, lipids, DNA, and facilitate the production of highly mutagenic metabolites [44]. Accordingly, oxidative stress plays a vital role in the tumorigenesis of PDAC [45]. Elevated ALP levels may serve as a potential predictor of oxidative stress [46], thereby contributing to cancer progression and poor disease outcomes [47, 48]. Because of this, patients with low AAPR might have a poor prognosis.

Herein, we demonstrated the prognostic value of AAPR for unresectable PDAC by multivariate and univariate models as well as PSM analyses. Besides, patients’ characteristics were well matched to minimize potential confounding bias. The prognostic value of AAPR was investigated in the entire cohort and further validated in the PSM cohort. Moreover, AAPR was confirmed to be associated with OS in most subgroups, which suggested that AAPR could be applied in patients with different clinicopathological features.

However, this study had a few limitations. First, our cohort was a retrospective cohort in a single center and composed of Chinese patients only; therefore, these results might not apply to other populations. Further multicenter prospective studies are thus recommended. Second, FOLFIRINOX is a standard regimen option for patients with metastasized PDAC. However, FOLFIRINOX is not widely used by Chinese patients because of its severe side effects. Whether AAPR could be used in patients receiving FOLFIRINOX remains unclear. Third, Karnofsky performance scale (KPS) [49] and tumor size [50] were reported as prognostic factors in advanced pancreatic cancer receiving chemotherapy and were not analyzed in this study. The potential confounding effect of these factors on AAPR needs further assessment. Forth, the underlying mechanism between the AAPR and cancer biology was not investigated and therefore needs further research.

In conclusion, our results suggest that pretreatment AAPR may serve as a prognostic predictor of overall survival in patients with unresectable PDAC. AAPR shows potential in improving the prediction performance of conventional prognostic model, which may help clinicians to identify patients at high risk and guide individualized treatment. Further prospective studies are needed to validate these results.