Prognostic role of pretreatment derived neutrophil to lymphocyte ratio in urological cancers.

[pjoshea13]

New meta-analysis below [1].

I reviewed the PCa on ... 3 years ago:

"Inflammation. [1] Neutrophil-to-Lymphocyte Ratio [NLR]" [2]

with a follow-up a year later:

"Neutrophil-to-Lymphocyte ratio [NLR] & Abiraterone." [3].

In the new meta-analysis, the purpose was to:

"... investigate the possible prognostic role of pretreatment derived neutrophil-lymphocyte ratio (dNLR) in urological cancers, including renal cell carcinoma (RCC), prostate cancer (PCa), and urothelial cancer (UCa)."

"A total of 12 studies embracing 6,585 patients were included in the meta-analysis."

"A higher dNLR was associated with an inferior biochemical recurrence-free survival ... and OS ... in PCa."

"A higher dNLR level ... could be an independent prognosis predictor in urological cancers." Yes, but inflammation can be treated. The NLR is a modifiable survival risk factor (IMO).

-Patrick

[1] ncbi.nlm.nih.gov/pubmed/317...

Int J Surg. 2019 Nov 7. pii: S1743-9191(19)30312-7. doi: 10.1016/j.ijsu.2019.10.043. [Epub ahead of print]

Prognostic role of pretreatment derived neutrophil to lymphocyte ratio in urological cancers: a systematic review and meta-analysis.

Su S1, Liu L2, Li C3, Zhang J1, Li S1.

Author information

1

Department of Urology, The NO.1 Hospital of Shijiazhuang, Shijiazhuang, China.

2

Department of Pathophysiology, Hebei Medical University, Shijiazhuang, China.

3

Department of Urology, The NO.1 Hospital of Shijiazhuang, Shijiazhuang, China. Electronic address: [e-mail].

Abstract

BACKGROUND:

To investigate the possible prognostic role of pretreatment derived neutrophil-lymphocyte ratio (dNLR) in urological cancers, including renal cell carcinoma (RCC), prostate cancer (PCa), and urothelial cancer (UCa).

MATERIALS AND METHODS:

Eligible studies were comprehensively searched from PubMed, Embase, Cochrane Library and Web of Science, up to April 2019. Pooled hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated to assess the relationships.

RESULTS:

A total of 12 studies embracing 6,585 patients were included in the meta-analysis. Our results indicated that a higher pretreatment dNLR was associated with a decreased cancer-specific survival (CSS, HR 2.67, 95% CI 1.06-6.71, P = 0.037) and disease-free survival (DFS, HR 2.02, 95% CI 1.03-3.94, P = 0.040) in RCC, but not for overall survival (OS, HR 1.05, 95% CI 0.71-1.53, P = 0.818). A higher dNLR was associated with an inferior biochemical recurrence-free survival (BRFS, HR 1.70, 95% CI 1.00-2.87, P = 0.049) and OS (HR 1.35, 95% CI 1.20-1.51, P < 0.001) in PCa. A higher dNLR was associated with a worse OS (HR 1.29, 95% CI 1.03-1.61, P = 0.029) and CSS (HR 1.51, 95% CI 1.06-2.15, P = 0.024) in UCa, but not for DFS (HR 1.44, 95% CI 0.89-2.34, P = 0.139).

CONCLUSION:

A higher dNLR level was negatively associated with OS, CSS, DFS and BRFS, forecasting that it could be an independent prognosis predictor in urological cancers.

Copyright © 2019. Published by Elsevier Ltd.

KEYWORDS:

derived neutrophil-lymphocyte ratio; meta-analysis; prognosis; urological cancer

PMID: 31707011 DOI: 10.1016/j.ijsu.2019.10.043

[2] healthunlocked.com/advanced...

[3] healthunlocked.com/advanced....

[snoraste]

No mention of threshold they used -

[pjoshea13 in reply to snoraste]

I found this 2018 study [1] to be of interest:

"in contrast to the majority of previous studies, we report the prognostic value of NLR as a continuous variable to avoid the loss of information inherent in data-driven categorization. We found that an increase in NLR of 1 standard deviation (SD) increased the risk of death by 35% (p<0.0001) after adjustment for other prognostic factors (ECOG-PS, age, gender, tumor type); we believe that “SD increase” may be a better indicator for decision purposes than binary cut-offs. In addition, we also compared the prognostic value of NLR with that of neutrophil or lymphocyte count alone. We found that an elevated neutrophil count significantly increased the risk of death by 26% whereas an elevated lymphocyte count decreased the risk of death by 20%, the magnitude of impact being lower than for NLR (35%)."

That is from the Discussion section:

"To the best of our knowledge, this is the first study with individual patient data to show that elevated NLR is associated with worse survival in patients with various metastatic solid tumors, irrespective of the primary tumor site, line of chemotherapy, age, gender and ECOG-PS. In order to account for the diversity of tumor types, stratification allowed NLR effect to be calculated first within each stratum before getting an overall result.

Current literature show that interest in NLR over the last decade has grown rapidly. An elevated NLR has been shown to be associated with reduced overall survival in patients with a variety of different individual tumor types [8, 15–26], disease stages (localized or metastatic) and treatments (chemotherapy or radiotherapy) [27]. However, very few previous trials have included patients with different tumor types [31, 32] as we did in this study. Inclusion of patients with a variety of tumor types allows us to investigate any differential prognostic value of NLR. We found that elevated NLR was associated with poor prognosis independently of tumor type. Thus, our results confirm data from a large meta-analysis suggesting that elevated NLR was a strong predictor of poor survival across all tumor types [28]. Because the meta-analysis was based on published data, the influence of other prognosis parameters such as ECOG-PS or age on the prognostic value of NLR was not reported. In addition, the influence of the line of chemotherapy was not assessed. In our study, we show that NLR prognostic value was not influenced by ECOG-PS, age, gender or tumor location. In addition, in another cohort of 134 patients treated with later lines of chemotherapy, we show that NLR prognostic value was still independently associated with worse survival. These results are in line with the fact that the prognostic value of NLR has been demonstrated in a wide range of disease stages, from localized disease to patients receiving palliative care [33] or terminally ill [34].

Furthermore, in contrast to the majority of previous studies, we report the prognostic value of NLR as a continuous variable to avoid the loss of information inherent in data-driven categorization. We found that an increase in NLR of 1 standard deviation (SD) increased the risk of death by 35% (p<0.0001) after adjustment for other prognostic factors (ECOG-PS, age, gender, tumor type); we believe that “SD increase” may be a better indicator for decision purposes than binary cut-offs. In addition, we also compared the prognostic value of NLR with that of neutrophil or lymphocyte count alone. We found that an elevated neutrophil count significantly increased the risk of death by 26% whereas an elevated lymphocyte count decreased the risk of death by 20%, the magnitude of impact being lower than for NLR (35%).

Published data suggest high variability in the optimal NLR cut-off value, ranging between 1.9 and 9.21 [28, 34]. In order to identify the optimal NLR cut-off value in our study, we performed extensive sensitivity analyses. We found that in patients with a variety of solid tumors, the “optimal” pre-treatment cut-off was between 3.5 and 4.5. Patients with an NLR ≥4.0 had a 53% increased risk of death (p<0.008, Fig 3B) compared with those who had an NLR value <4.0, after adjustment for age, gender, disease site, and Eastern Co-operative Oncology Group performance status. In the largest meta-analysis, median cut-off for high NLR was 4.0 (range 1.9–7.2) and patients with NLR above this cut-off had an 81% increased risk of death (p<0.001) [28], a value that is similar to our findings. Nevertheless, the optimal NLR cut-off was highly variable based on the time at which survival was assessed and the indices used for optimality assessment, questioning the real-life practical “decisional value” of such a cut-off.

Our study is probably the first to show that categorizing the NLR resulted in a significant time-interaction and that introducing a time-dependent term in the model could modify the NLR effect. Indeed, NLR effect was even more pronounced after introducing time-dependent term suggesting that this should always be included in future investigations and its impact on the prognostic value of NLR discussed in studies using an arbitrary cut-off. Moreover we report again for the first time that the index used for optimality assessment of the time-dependent NLR cut-off had an impact on the resulting optimal value. We found that the log-rank method was the most stable across time and gave an optimal value between 4.0 and 4.5. Given our results on the variability of NLR effect, we believe that no standardized cut-off values may be reliable enough to be used in the clinics for decision purposes.

Mechanisms underlying the prognostic value of NLR remain unclear. It is widely considered that NLR reflects a balance between inflammatory status, represented by neutrophils, and adaptive immunity, represented by lymphocytes [9]. Thus, an increase in NLR could reflect an increase in inflammation status or a decrease in anti-tumor immunity, or both. Neutrophils have been shown to produce pro-inflammatory cytokines such as interleukin (IL)-6, which is a major mediator of the hepatic acute-phase response and induces production of acute-phase proteins including CRP, a well-known circulating marker of inflammation. High CRP has also been reported to predict poor prognosis in many tumor types [14]. Consistent with these observations, inflammation is a well-known indicator of tumor progression [9].

A high density of tumor-infiltrating lymphocytes (TIL) has been shown to be strongly associated with a good prognosis in the majority of solid tumors [35]. Nevertheless, no correlation has been documented between density of TIL and circulating lymphocyte count. In fact, the example of renal cell carcinoma where a high density of CD8+ TIL [35] and a low circulating lymphocyte count [36] are associated with a poor prognostic shows the opposite. Few studies have specifically analyzed circulating leucocytes in patients with cancer. Ohki et al. used flow cytometry to characterize peripheral blood mononuclear cells (PBMC) from patients with various tumor types compared with healthy donors [37]. They showed a significant increase in the percentage of myeloid-derived suppressor cells (MDSCs) in PBMC from patients versus healthy donors. In addition, they reported a strong correlation between circulating MDSC and NLR.

Important strengths of our study are its prospective collection of clinical and biological data thanks to the PROCHE program, which had already demonstrated its usefulness [29]. Another strength is the inclusion of patients with a wide range of different solid tumors, which increases the clinical applicability of our results.

Our study has some limitations that should be noted. Firstly, this is a single center study. Nevertheless this point strengthens our results because the management of all patients was highly homogenous and we are sure that data collected through the PROCHE program are reliable. Secondly, we did not look at comorbidities and concomitant medications meaning that any influence of these factors on hematologic variables could not be taken into account. Nevertheless, we verified that no patient had received any growth factor treatment and/or any corticosteroid at the time of NLR collection, which could have an obvious influence on the NLR value. Thirdly, NLR was not assessed during cancer treatment; therefore, data on changes in this biomarker over time and their predictive value for response to therapy could not be determined. Some studies have suggested that the early normalization of the NLR during chemotherapy was a surrogate marker for better survival [38–40].

In conclusion, the results of this study add to knowledge about NLR as an indicator of mortality risk in patients with cancer, providing better understanding of the consistency and magnitude of the prognostic information given by this biomarker. There is strong evidence that the increased mortality risk associated with high NLR is independent of the primary tumor site, line of chemotherapy, age, gender and ECOG-PS. Therefore, NLR alone might be used for prognosis across a wide range of solid tumor types in clinical practice. Nevertheless, challenges discussed above regarding the use of any arbitrary cut-off restrict its use. To overcome this limit and to move forward we are convinced that we have to monitor the variation of the NLR during systemic therapy in order to define patients who will not respond."

***

-Patrick

[1] ncbi.nlm.nih.gov/pmc/articl...

[snoraste in reply to pjoshea13]

Thanks Patrick -

I read thru the article, and they do have a point. Real life is far more continuous than binary (if the threshold is 4 for NLR, someone with 3.9 has far better survival odds than one with 4.1!). There's real information added there.

I wish they mentioned the "mean" and the "Standard Deviation" of NLRs so we can run some real numbers.

[Garp41]

Patrick.......you said

The NLR is a modifiable survival risk factor (IMO)

How would you do this?

Doug

[pjoshea13 in reply to Garp41]

This is what I wrote 3 years ago:

healthunlocked.com/advanced...

-Patrick

[Garp41 in reply to pjoshea13]

Thanks Patrick....I take most of them.

[jdm3 in reply to pjoshea13]

Thank you. I went back and re-read this post. Great information and summary with good references. If there was a hall of fame for PCa researchers, you would be right there with the best of them.

[LearnAll]

I have been keeping a chart of my NLR and PLR. When I was first diagnosed with severe UTI and PCA my NLRatio was 5.7 but after treatment with ADT and antibiotics, and anti inflammatory foods and supplements my NL ratio dropped down to 0.92 and is staying around 1 to 2.

Every time I have urinary symptom, NLR goes up and then falls down after treatment. And same exact pattern in chart showing PL ratio (platelet lymphocyte ratio)

I do find this correlation but how this relates to prognosis ? If any one there who can explain.

[PhilipSZacarias]

Hello Patrick, Your post is quite timely - I will read the full paper over the next day or two. I have been tracking NLR (neutrophil/lymphocyte ratio), PLR (platelet/lymphocycte ratio) and SII (Systematic Inflammation Index = P * N/L) for the last couple of years. The PSA has been rising slowly since March this year (from 0.085 to 0.17; tests are taken 1-1.5 months apart). I found a positive correlation between the PSA rise and rises in NLR and SII (correlation coefficients of 0.8 and 0.945, respectively). Needless to say, I am working on a change to my regimen. Cheers, Phil