Prognostic significance of HALP (hemoglobin, albumin, lymphocyte and platelet) in patients with bladder cancer after radical cystectomy

The outcome of bladder cancer after radical cystectomy is heterogeneous. We aim to evaluate the prognostic value of HALP (hemoglobin, albumin, lymphocyte and platelet) and explore novel prognostic indexes for patients with bladder cancer after radical cystectomy. In this retrospective study, 516 patients with bladder cancer after radical cystectomy were included. The median follow-up was 37 months (2 to 99 mo). Risk factors of decreased overall survival were older age, high TNM stage, high American Society of Anesthesiologists (ASA) grade and low HALP score. The predictive accuracy was better with HALP-based nomogram than TNM stage (C- index 0.76 ± 0.039 vs. 0.708 ± 0.041). By combining ASA grade and HALP, we created a novel index—HALPA score and found it an independent risk factor for decreased survival (HALPA score = 1, HR 1.624, 95% CI 1.139–2.314, P = 0.007; HALPA score = 2, HR 3.471, 95% CI: 1.861–6.472, P < 0.001).The present study identified the prognostic value of HALP and provided a novel index HALPA score for bladder cancer after radical cystectomy.

HALP-based risk model for bladder cancer patients after radical cystectomy. To further assess the prognostic ability of HALP and other variables, we used nomogram including the independent risk factors in the multivariable regression analysis (Fig. 3). Predicted 3-and 5-year survival was similar to the actual rates (Fig. 4). The predictive accuracy of C-indices was 0.76 ± 0.039 for the HALP-based nomogram as compared with 0.708 ± 0.041 for the TNM stage-based nomogram.
Prognostic value of novel index HALPA. In the multivariate analysis and nomograms, ASA was a significant factor in addition to TNM stage and HALP score. Therefore, we combined ASA grade and HALP score and created a new index, HALPA score. The score of HALPA was determined as follows: HALPA = 0 (HALP ≥ 22.2 and ASA grade = 1&2); HALPA = 1 (HALP < 22.2 or ASA grade = 3&4); and HALPA = 2 (HALP < 22.2 and ASA grade = 3&4). Therefore, a higher HALPA score indicated higher risk of decreased survival. Univariate and multivariable analyses were performed again with HALPA score. On log-rank test, HALPA score was a significant indicator of poor survival for all patients and NMIBC or MIBC patients (Fig. 5). Notably, no NMIBC patient had a HALPA score of 2, which also showed the ability of HALPA score to distinguish the tumor stage of bladder cancer patients. On multivariable analysis, high HALPA score remained an independent risk factor of decreased survival along with older age and TNM stage (HALPA score = 1, HR = 1.624, 95% CI:1.139-2.314, P = 0.007; HALPA score = 2, HR = 3.471, 95% CI: 1.861-6.472, P < 0.001) ( Table 4).

Discussion
Besides traditional TNM stage and histologic classification, several haematological-based indices such as hemoglobin, albumin, and CRP levels, NLR, PLR and LMR were found associated with outcome with bladder cancer 7,10,[12][13][14]17 . In the present study, we evaluated the prognostic value of a recently reported index, HALP, which combined hemoglobin and albumin levels and lymphocyte and platelet count, and found it as a good prognostic index for overall survival of patients with bladder cancer after radical cystectomy. By combining ASA grade and HALP score, we also created HALPA score and found it as an independent risk factor of decreased overall survival. Cancer-related anemia is a common complication for cancer patients because of chromic blood loss, iron deficiency, vitamin deficiency and inflammation imbalance in terms of increased expression of interleukin and tumor necrosis factor. Jung et al. 7 reported that 81 of 200 (40.5%) bladder cancer patients undergoing radical cystectomy had preoperative anemia and that anemia was significantly associated with disease recurrence and cancer-specific mortality and for MIBC patients, 53.3% patients were anemic, and older age, neoadjuvant chemotherapy, female sex, and low BMI were independent predictors of preoperative anemia. In this study, anemia was present in 27.9% patients (12.4% NMIBC, 37.5% MIBC, P < 0.001), which was lower than in previous reports. This difference may be due to use of different population. On univariate analysis, anemia was significantly associated with decreased overall survival (P < 0.001).
Hypoalbuminemia is another nutrition deficiency index. Lambert et al. 11 reported that for bladder cancer patients after radical cystectomy, 16.5% had low-albumin level (<3.5 g/dL), and low albumin level was significantly associated with increased risk of mortality (HR = 1.76, P = 0.04). Another study found that 197 of 1471 (13.4%) bladder cancer patients had a low serum albumin level, and low albumin level was independently associated with reduced recurrence-free survival (HR 1.68, P = 0.006) and overall survival (HR 1.93, P < 0.001). Moreover, the authors also found low serum albumin associated with higher 90-day complication rate (42% vs. 34%, P = 0.03) and 90-day mortality (7.6% vs. 3.3%, P = 0.008) 17 . These findings document the adverse effects of nutritional deficiency on survival of bladder cancer after radical cystectomy and demonstrate the importance of nutritional interventions.
Tumor-promoting inflammation is one of the hallmarks of cancer 18 . For example, neutrophils and platelets could promote carcinogenesis, angiogenesis, invasion, or metastasis by secreting proinflammatory cytokines 19,20 .    However, lymphocytes could inhibit tumor proliferation and migration via cytotoxicity 21 . Therefore, high PLR level, which reflects high platelet count and low lymphocyte count was found associated with overall survival in patients with bladder cancer undergoing radical cystectomy 14 .
The association of ASA score and long-term survival outcomes of bladder cancer patients after radical cystectomy is rarely reported. One study found that 740 of 1964 (64.8%) patients who underwent radical cystectomy had a high ASA score (3 or 4), and a high ASA grade remained independently associated with decreased overall survival (HR 1.45, 95% CI: 1.13-1.86, P = 0.003) 17 . In this study, a high ASA score was also significant independent risk factor for decreased overall survival and the nomogram including ASA grade and HALP score had better predictive accuracy than TNM stage. By combining ASA grade and HALP score, we created a novel index, HALPA score, and found it to have a specific distribution in patients at different stages and a significant independent risk factor with higher risk ratio than HALP score or ASA alone.
There are several limitations to this study. First, the retrospective nature has inherent limitations. Second, this was a single-center study and could not avoid the bias of population choice. Third, the use of cutoffs for continuous variables might weaken the reliability of the risk system. Finally, ASA is a subjective measure and this may limit the generalizability and overall utility. The prognostic ability and accuracy of HALP and HALPA score for bladder cancer patients warrants prospective multi-center validation.

Conclusions
HALP may be a good prognostic index for overall survival for patients with bladder cancer after radical cystectomy. Combining ASA grade and HALP score in the HALPA score could distinguish tumor stage and was an

Data collection. This study was approved by the Institutional Review Board of Peking University First
Hospital. Between 2006 and 2012, 516 bladder cancer consecutive patients underwent radical cystectomy in the Department of Urology, Peking University First Hospital. This research was carried out in accordance with the approved guidelines and informed consent was obtained from all patients. Hematological features including hemoglobin and albumin levels and blood cell counts were preoperative collected within 3 days before surgery. Other clinicopathological data included gender, age, smoking history, hypertension history, histology subtype, grade, TNM stage and American Society of Anesthesiologists (ASA) grade.
The follow-up of bladder cancer patients after radical cystectomy was according to the routine procedure in our institution. The period from the operation date until the time of death resulting from any cause was calculated as overall survival time. HALP was calculated as hemoglobin (g/L) × albumin (g/L) levels × lymphocyte count   (/L)/platelet count (/L) and PLR as platelet count/lymphocyte count. Anemia was defined according to preoperative hemoglobin level (males <120 g/L and females <110 g/L) and hypoalbuminemia according to preoperative albumin level (<35 g/L).
Statistical analysis. X-tile software v3.6.1 (Yale University) was used to determine the cut-off values of PLR and HALP 22 . SPSS v22.0 (SPSS Inc., Chicago, IL, USA) was used for all statistical analyses. Chi-square test was used to evaluate the association between clinicopathological data and HALP. Overall survival was estimated by the Kaplan-Meier method and the log-rank test was used for univariate analysis. Variables with significant differences on univariate analysis (P < 0.05) were then included in the multivariable survival analyses with a Cox proportional-hazards regression model, estimating hazard ratios (HRs) and 95% confidence intervals (CIs). Nomogram and production of calibration curves involved use of R v3.3.2 (R Foundation for Statistical Computing) with the rms package (Regression Modeling Strategies). A two-tail P < 0.05 was considered statistically significant.