Preoperative Thrombocytosis is Not Associated with Overall Survival in 309 Glioblastoma Patients

Keywords

platelet count - overall survival - glioblastoma - aspirin - hemoglobin

Introduction

Early citations on the interaction between the hemostatic system and cancer progression go back to the 19th century and are nowadays put into perspective of newer insights and a broader scientific background.[1] In recent years, an association of thrombocytosis with a worse prognosis was shown for many solid tumors such as from the lung, colon, breast, gastric, ovarian, and melanoma.[2] Glioblastoma multiforme (GBM) depicts a primary brain tumor with only poor survival and only limited therapeutic options. Best mean survival is known for the adjuvant radiotherapy and chemotherapy with temozolomide (TMZ) after the Stupp protocol.[3] In 2007, Brockmann et al have shown an association of thrombocytosis with decreased overall survival in 158 patients diagnosed with GBM.[4] Another study could not validate this association in 140 patients with GBM.[5] Beside the platelet count, the mean platelet volume (MPV) was already identified as marker for tumor progression. For example, the MPV was shown to be either increased in patients with gastric cancer and hepatocellular carcinoma or decreased in patients with non–small cell lung carcinoma.[6] [7] [8] Furthermore, the alteration of the MPV was identified as a prognostic factor for the overall survival in bladder cancer.[9]

The main aim of the present study was to validate the findings of Brockmann et al[4] on platelet count and its association with overall survival in a cohort of 309 GBM patients as well as to investigate if there is an association between the MPV and overall survival. Furthermore, other available patient data as well as hemostatic and hematologic parameters were investigated to see if there is an association with the overall survival in GBM patients.

Methods

All patients with a first-diagnosed and histopathologically proven GBM between 2005 and 2015 were included in the study. Individuals with a secondary GBM (n = 34) or missing preoperative laboratory investigations (n = 1) were excluded. Data analysis of each patient included gender, age, laboratory investigation prior to surgery (platelet count, platelet volume, leucocyte count, hemoglobin, activated partial thromboplastin time [aPTT], fibrinogen level), Karnofsky performance score, tumor location, tumor volume, extent of tumor resection, adjuvant therapy, postoperative survival time, cardiovascular comorbidities, and co-medication. Preoperative tumor volume was assessed with a layer-by-layer based segmentation and calculation of the total volume on preoperative magnetic resonance imaging (MRI) using OsiriX (Pixmeo SARL, Bernex, Switzerland). Extent of tumor resection based on the postoperative MRI was classified as total resection, subtotal resection, and tumor biopsy. In some patients (n = 49), no postoperative MRI was available and the estimation of the extent of resection was based on the postoperative cerebral computertomography (CCT) and the surgeon's intraoperative impression. Although this is more inaccurate than a postoperative MRI, the subdivision of total resection, subtotal resection, and tumor biopsy is feasible by this approach. Data for overall survival were obtained by phone calls with the families of the patients or the general practitioners. Thrombocytosis was defined as platelet count greater than 400 Gpt/L. Anemia was defined as hemoglobin less than 7.5 mmol/L.[10]

The study was approved by the local ethics committee of our university (BB089/08b). All patients gave their informed consent to participate in the study. All analyses were performed in accordance with relevant guidelines and regulations.

Statistical Analysis

Descriptive statistics are presented as counts and percentages for categorical and ordinal data. For continuous data, means, standard deviations, and the range of values are presented. The association between potential risk factors and control variables was assessed using the Kaplan–Meier survival analysis and cox regression. A p value ≤ 0.05 was considered statistically significant.

Results

Patient Characteristics

In this study, 309 patients with a first-diagnosed primary GBM were included (126 females, 183 males, mean age of 65.9 years, and range of 27 to 89 years). Mean overall survival was 14.5 months. In all, 283 (91.6%) patients died between 0.03 and 97.4 months after surgery. Twenty-six patients are still alive. The mean preoperative tumor volume was 36.1 cm³ (range: 0.2–140.4 cm³) in 268 patients. In 41 patients, the preoperative tumor volume was not obtainable. The tumor was located in the frontal lobe (n = 70), the temporal lobe (n = 86), the frontotemporal lobe (n = 9), the parieto-occipital region (n = 72), the central region (n = 7), the corpus callosum (n = 30), the insula and basal ganglia (n = 23), the cerebellum (n = 3), purely intraventricular (n = 2), or in the pons (n = 2). One hundred and thirty and 116 patients got a gross total and subtotal tumor resection, respectively. Sixty-three patients got a tumor biopsy. None of the patients who underwent a tumor biopsy received tumor resection during further course of the disease. The mean Karnofsky index obtained at the day of discharge after surgery was 80 (range: 0–100%).

In all, 155/309 patients received the standard adjuvant treatment according to Stupp et al.[3] Of these, 18 patients were continued with TMZ therapy thereafter, whereas 37/155 patients had another chemotherapy after the Stupp protocol. Nine of 155 patients continued with TMZ and another chemotherapy after the Stupp protocol, whereas 69/309 patients were treated with radiotherapy alone as adjuvant treatment and 5 patients got TMZ alone as adjuvant treatment. 59/309 patients received no adjuvant therapy. No data regarding the specific postinterventional therapy were available for 21/309 patients.

The mean platelet count in 309 patients was 255 Gpt/L (range: 56–623 Gpt/L). Only 14 (4.5%) patients had a platelet count greater than 400 Gpt/L and can be considered as thrombocytosis. The MPV in 309 patients was 10.4 fL (range: 7.4–13.6 fL). The mean hemoglobin level in 309 patients was 8.9 mmol/L (range: 5.4–11.3 mmol/L). The mean total white blood cell count in 309 patients was 11.59 Gpt/L (range: 2.84–30.2 Gpt/L). aPTT was available from 298 patients with a mean of 22 seconds (range: 17–39 seconds). The fibrinogen level was obtainable in 101 patients with a mean of 3.1 g/L (range: 1.4–5.8 g/L).

Fifty-eight 58 (18.7%) patients had acetylsalicylic acid (aspirin) as co-medication. Hereof 26 and 32 patients got aspirin as primary and secondary prophylaxis, respectively.

Association of Patient Age, Preoperative Tumor Volume, Tumor Location, Karnofsky Performance Score, Extent of Resection, and Adjuvant Therapy with Overall Survival of GBM Patients

The study cohort showed a statistically significant association of a decreased overall survival of the patients with a higher patient age (p < 0.001), a higher preoperative tumor volume (p = 0.013), and a lower Karnofsky performance score (p < 0.001; [Fig. 1A, B, D]; [Table 1]). A statistically significant association was seen between overall survival of the patients with the localization of the tumor (p = 0.007). GBM in the corpus callosum had a significantly worse overall survival than GBM in the frontal lobe (p < 0.001). Resection status was shown to be important for overall survival (p < 0.001; [Fig. 1C]; [Table 1]). Patients with a gross total tumor resection had an increased overall survival compared with patients with a subtotal tumor resection (p = 0.001) or a tumor biopsy (p < 0.001). A statistically significant association was seen between overall survival of the patients and different adjuvant treatment modalities (p < 0.001). Patients without any adjuvant therapy or with radiotherapy alone had a decreased overall survival compared with patients who received the standard treatment according to Stupp (p < 0.001). Patients with TMZ monotherapy or TMZ monotherapy combined with another chemotherapy after the Stupp protocol had an increased overall survival compared with patients who received the Stupp protocol alone (p < 0.001 and p = 0.006, respectively).

Fig. 1 (A) Association of patient age, (B) Karnofsky performance score, (C) extent of tumor resection, and (D) preoperative tumor volume with the overall survival of glioblastoma multiforme (GBM) patients. The differences between the groups were assessed using the Kaplan–Meier survival analysis.

Table 1 Median survival time by age, preoperative tumor volume, Karnofsky Performance Status, extent of tumor resection and platelet count

Predictor	Category	N	Median survival time (d)	95% confidence interval (CI)	p value[a]
Age	≤70 y	184	365	281–449	<0.001
	>70 y	125	153	115–191
Preoperative tumor volume	≤80 cm3	242	287	228–346	0.001
	>80 cm3	26	137	86–188
Karnofsky Performance Status	≤70	104	118	82–154	<0.001
	>70	203	365	322–408
Extent of tumor resection	Total	130	423	363–483	<0.001
	Subtotal	116	277	185–369
	Biopsy	63	67	55–79
Platelet Count (Gpt/l)	<350	278	254	198–310	0.413
	350–399	17	283	269–297
	400–450	10	399	303–495
	>450	4	131	0–320

^a Log-rank test.

Association of Platelet Count and Mean Platelet Volume with the Overall Survival of GBM Patients

In the study cohort, no difference in overall survival could be observed between patients with thrombocytosis (n = 14) and those without thrombocytosis (n = 295; p = 0.586; [Fig. 2]). Furthermore, there was no difference in overall survival between patients with platelet counts of <350 (n = 278), 350 to 399 (n = 17), 400 to 449 (n = 10), and >450 (n = 4; p = 0.413; [Fig. 3]; [Table 1]). Cox regression models including potential confounders have shown that the statistically insignificant association of overall survival with platelet count was not affected by the inclusion of patient age (the p value for the effect of platelet count ≥400 = 0.789), preoperative tumor volume (p = 0.105), Karnofsky performance status (p = 0.244), tumor localization (p = 0.147), and extent of tumor resection (p = 0.436). The patients with or without thrombocytosis did not show any statistically significant differences, except the preoperative platelet and leucocyte count ([Table 2]).

Fig. 2 Association of a preoperative thrombocytosis with the overall survival in of glioblastoma multiforme (GBM) patients. The difference between the groups (platelet count greater than or less than 400 Gpt/l) was assessed using the Kaplan–Meier survival analysis.

Fig. 3 Association of the preoperative platelet count with the overall survival in of glioblastoma multiforme (GBM) patients. The differences between the groups were assessed using the Kaplan–Meier survival analysis.

Table 2 Comparison of patients with a preoperative thrombocytosis (>400 Gpt/l, n = 14) and patients without (<400 Gpt/l, n = 295)

		<400 Gpt/l	>400 Gpt/L	Statistical test	p value
	Age	66.1 (28–89)	61.5 (27–76)	t-test	0.13
Gender	Female	118	8	Fisher exact	0.27
	Male	177	6
Laboratory	Tumor volume	29.3 (0.2–140.4)	54 (17.8–98.1)	t-test	0.09
	Platelet count	248 (56–393)	434 (404–623)	t-test	0.0001
	Platelet volume	10.4 (7.4–13.6)	10.05 (8.9–10.9)	t-test	0.06
	Leucocyte count	17.2 (7.7–30.2)	10.7 (2.8–28.7)	t-test	0.0001
	fibrinogen	2.6 (2–4.3)	2.9 (1.4–6.1)	t-test	0.51
	aPTT	22 (17–25)	22 (17–39)	t-test	0.26
	hemoglobin	8.5 (6.5–9.8)	8.9 (5.4–11.3)	t-test	0.12
	KPS	90 (0–100)	80 (0–100)	t-test	0.49
EOR	Biopsy only	61	2	Chi-squared	0.99
	STR	109	7
	GTR	125	5
adjuvant treatment	none	58	1	Chi-squared	0.09
	RTX alone	69	2
	Stupp	155	11

Abbreviations: aPTT, activated partial thromboplastin time; EOR, extent of resection; GTR, gross total resection; KPS, Karnofsky Performance Status; RTX, radiotherapy; STR, subtotal resection.

In the study cohort, no association could be observed between the overall survival of patients and the preoperative MPV (p = 0.739). The covariate analyses including potential confounders have shown that the statistically insignificant association of overall survival with MPV was not affected by the inclusion of patient age (p value for the effect of MPV was 0.824), preoperative tumor volume (p = 0.554), Karnofsky performance status (p = 0.221), tumor localization (p = 0.907), and extent of tumor resection (p = 0.816).

Association of Leucocyte Count and Hemoglobin Level, Fibrinogen Level, and aPTT with the Overall Survival of GBM Patients

The study cohort showed no association between the overall survival of patients and the preoperative leucocyte count (p = 0.495). Furthermore, there was no difference in the overall survival of patients with a leucocyte count greater than 11 Gpt/L (p = 0.741). In the study cohort, no association could be found between the overall survival of the patients and the preoperative aPTT (p = 0.955) or the fibrinogen level (p = 0.751).

In the study cohort, a trend of an association between the overall survival of patients and the preoperative hemoglobin level could be observed (p = 0.062). Furthermore, the overall survival of patients with a hemoglobin level below 7.5 mmol/L was decreased (p = 0.016; [Fig. 4]). Including the patient age and the Karnofsky performance status as covariates in this analysis, the statistically significant association between hemoglobin below 7.5 mmol/L and the overall survival in GBM patients persists (p = 0.03 and 0.04, respectively; [Table 3]). Furthermore, it could be shown that the association of hemoglobin as continuous marker and the overall survival becomes statistically significant if the Karnofsky performance score is considered a covariate (p = 0.04; [Table 3]).

Fig. 4 Association of the preoperative hemoglobin level with the overall survival in of glioblastoma multiforme (GBM) patients. The differences between the groups were assessed using the Kaplan–Meier survival analysis.

Table 3 Covariate analysis for the association of hemoglobin with overall survival of GBM patients

Hemoglobin	Covariables	HR (hemoglobin)	p value (hemoglobin)
</>7.5 mmol/L		0.52	0.016
</>7.5 mmol/L	Age	0.54	0.027
</>7.5 mmol/L	Karnofsky Performance Status	0.57	0.043
continuous		0.87	0.062
continuous	Age	0.93	0.327
continuous	Karnofsky Performance Status	0.86	0.038

Abbreviation: GBM, glioblastoma multiforme; HR, hazard ratio.

Association of Co-medication of Aspirin with Overall Survival in GBM Patients

In the study cohort, a statistically significant decreased overall survival was seen in patients who received aspirin as co-medication (p = 0.034). This observation was most pronounced in patients who got aspirin as primary prophylaxis (p = 0.031). However, in a further analysis with patient age as a covariate, this effect was not seen any more (p = 0.674). In the study cohort, no difference in the overall survival could be obtained between patients who had clopidogrel as co-medication and those who did not (p = 0.097).

Discussion

Summary of the Key Results

The present study with 309 first-diagnosed primary GBM patients could validate well-known risk factors of a decreased overall survival like higher patient age, a larger preoperative tumor volume, Karnofsky performance status, extent of resection, tumor localization, and adjuvant treatment. No association between the overall survival of GBM patients and the preoperative platelet count, MPV, leucocyte count, aPTT, fibrinogen level, or aspirin co-medication was observed. Patients with a preoperative hemoglobin below 7.5 mmol/L had a decreased overall survival.

Limitations of the Study

This is a retrospective study. Thus, it has a low class of evidence. Nevertheless, all the studies addressing the association of thrombocytosis with overall survival in GBM patients to date are retrospective as well. The present study could validate well-known risk factors for a decreased overall survival in GBM patients as higher patient age, larger preoperative tumor volume, tumor location, extent of resection, Karnofsky performance score, and the kind of the adjuvant therapy after surgery. Thus, the present study cohort of 309 GBM patients is comparable to other cohorts.

The present study did only analyze the platelet count and other hematologic parameters prior to any specific therapy to the patient. This time point was chosen because it is the only well-standardized time point. In the further course of the disease, many factors have an impact on platelet count and other hematologic markers such as surgery and adjuvant therapy with radiation and TMZ.[11] [12]

Because the MGMT methylation status began to be assessed on a regular basis in our department only since 2013, it could not be included in the analyses of the current study. However, it is important to do so in future studies, because the MGMT methylation status is important to predict the responsiveness to alkylating chemotherapies in GBM.[13]

Association of Platelet Count and Mean Platelet Volume with Overall Survival in GBM Patients

The importance of the association between the hemostatic system and cancer progression is well known and excellently reviewed by other groups.[1] [2] [14] [15] Many solid tumors are well known for an association between thrombocytosis and decreased overall survival of the patients.[2] The literature regarding the association between platelet count and overall survival in GBM patients is incongruent. Brockmann et al could show in a cohort of 158 GBM patients that preoperative thrombocytosis is associated with a decreased overall survival.[4] Lopes et al did not find any association between the platelet count and overall survival in a cohort of 140 GBM patients.[5] Another study could show that a drop of the platelet count after adjuvant therapy in GBM patients is a good prognostic factor.[16] The present study could not find any association between the preoperative platelet count and the overall survival in a cohort of 309 GBM patients. Interestingly, in the present study, only 14 of 309 patients (4.5%) presented with thrombocytosis in the preoperative laboratory investigation, compared with 29 of 153 patients (19%) in the study of Brockmann et al.[4] The reason for this difference remains elusive.

Furthermore, the present study could not reveal an alteration of the MPV or an association between the MPV and the overall survival in GBM patients. Recently it was shown that circulating platelets in GBM patients have an increased activation status, which was assessed by the quantification of several surface antigens and the release of sphingosine-1-phosphate.[17] Alteration of the MPV would be an indirect sign for increased platelet activation. However, alteration of the MPV usually occurs in a status of chronic inflammation. The GBM microenvironment and the circulation of GBM patients indeed are immunocompromised and anti-inflammatory,[18] [19] [20] which can be one possible explanation for a missing alteration of the MPV in these patients.

One main difference between GBM and malignancies outside of the central nervous system (CNS) is that GBM patients virtually never suffer from metastasis outside of the CNS. Nevertheless, many pieces of evidence suggest that GBM indeed is present outside of the CNS: (1) peripheral circulating tumor cells can be detected in GBM patients; (2) circulating platelets can contain GBM EGFRvIII RNA; and (3) donor organs from GBM patients have the risk of developing an extracranial metastasis in the recipient patient.[21] [22] [23] Platelets are well known to play an immanent role in the formation of metastasis.[24] [25]

Association of Leucocyte Count and Hemoglobin Level, Fibrinogen Level, and aPTT with the Overall Survival of GBM Patients

The present study did not find an association between the overall survival in GBM patients and the preoperative fibrinogen level in the plasma of the patients, the aPTT, and the total white blood cell count. Since a differentiation of white blood cells is not standard in our preoperative assessment, we cannot provide data for the neutrophil-to-lymphocyte (N/L) or the platelet-to-lymphocyte (P/L) ratio. A high N/L ratio was described to be associated with poor overall survival in GBM patients.[26] Nevertheless, another study could not confirm these findings.[5]

The present study shows an association between decreased overall survival in GBM patients and low preoperative hemoglobin level, which had also been shown by another group.[27] This finding in this study was not biased by the age of the patients.

Association of Co-medication with Aspirin to Overall Survival in GBM Patients

The present study did not find an association between overall survival of GBM patients and intake of aspirin, which is in line with the results of the largest prospective epidemiologic study with over 300,000 participants.[28] In this study, no association was found between daily intake of aspirin or other nonsteroidal anti-inflammatory drugs (NSAIDS) and the development of glioma and/or GBM. Furthermore, Brockmann et al could show, in animal experiments, that GBM tumor growth is not inhibited in platelet-depleted mice, suggesting a minor direct growth promoting effect of platelets in this model.[29]

Conclusion

The present study did not find any association between preoperative platelet count and overall survival in 309 patients with GBM. This is the largest series to date assessing that topic. In addition, preoperative hemoglobin level below 7.5 mmol/L is associated with a decreased overall survival in the present study.

• References

• 1 Buller HR, van Doormaal FF, van Sluis GL, Kamphuisen PW. Cancer and thrombosis: from molecular mechanisms to clinical presentations. J Thromb Haemost 2007; 5 (Suppl. 01) 246-254
• 2 Buergy D, Wenz F, Groden C, Brockmann MA. Tumor-platelet interaction in solid tumors. Int J Cancer 2012; 130 (12) 2747-2760
• 3 Stupp R, Mason WP, van den Bent MJ. et al; European Organisation for Research and Treatment of Cancer Brain Tumor and Radiotherapy Groups, National Cancer Institute of Canada Clinical Trials Group. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005; 352 (10) 987-996
• 4 Brockmann MA, Giese A, Mueller K. et al. Preoperative thrombocytosis predicts poor survival in patients with glioblastoma. Neuro-oncol 2007; 9 (03) 335-342
• 5 Lopes M, Carvalho B, Vaz R, Linhares P. Influence of neutrophil-lymphocyte ratio in prognosis of glioblastoma multiforme. J Neurooncol 2018; 136 (01) 173-180
• 6 Cho SY, Yang JJ, You E. et al. Mean platelet volume/platelet count ratio in hepatocellular carcinoma. Platelets 2013; 24 (05) 375-377
• 7 Inagaki N, Kibata K, Tamaki T, Shimizu T, Nomura S. Prognostic impact of the mean platelet volume/platelet count ratio in terms of survival in advanced non-small cell lung cancer. Lung Cancer 2014; 83 (01) 97-101
• 8 Osada J, Rusak M, Kamocki Z, Dabrowska MI, Kedra B. Platelet activation in patients with advanced gastric cancer. Neoplasma 2010; 57 (02) 145-150
• 9 Wang X, Cui MM, Xu Y. et al. Decreased mean platelet volume predicts poor prognosis in invasive bladder cancer. Oncotarget 2017; 8 (40) 68115-68122
• 10 Cappellini MD, Motta I. Anemia in clinical practice-definition and classification: does hemoglobin change with aging?. Semin Hematol 2015; 52 (04) 261-269
• 11 Lombardi G, Rumiato E, Bertorelle R. et al. Clinical and genetic factors associated with severe hematological toxicity in glioblastoma patients during radiation plus temozolomide treatment: a prospective study. Am J Clin Oncol 2015; 38 (05) 514-519
• 12 Zhang F, Guo X, Xing B, Yang Y, Xu Z. Hepatic and renal functions and blood cell counts in brain tumor patients during the perioperative period. J Clin Neurosci 2019; 69: 190-197
• 13 Wick W, Weller M, van den Bent M. et al. MGMT testing: the challenges for biomarker-based glioma treatment. Nat Rev Neurol 2014; 10 (07) 372-385
• 14 Goubran HA, Burnouf T, Radosevic M, El-Ekiaby M. The platelet-cancer loop. Eur J Intern Med 2013; 24 (05) 393-400
• 15 Jain S, Harris J, Ware J. Platelets: linking hemostasis and cancer. Arterioscler Thromb Vasc Biol 2010; 30 (12) 2362-2367
• 16 Williams M, Liu ZW, Woolf D. et al. Change in platelet levels during radiotherapy with concurrent and adjuvant temozolomide for the treatment of glioblastoma: a novel prognostic factor for survival. J Cancer Res Clin Oncol 2012; 138 (10) 1683-1688
• 17 Marx S, Splittstöhser M, Kinnen F. et al. Platelet activation parameters and platelet-leucocyte-conjugate formation in glioblastoma multiforme patients. Oncotarget 2018; 9 (40) 25860-25876
• 18 Bloch O, Crane CA, Kaur R, Safaee M, Rutkowski MJ, Parsa AT. Gliomas promote immunosuppression through induction of B7-H1 expression in tumor-associated macrophages. Clin Cancer Res 2013; 19 (12) 3165-3175
• 19 Markovic DS, Vinnakota K, Chirasani S. et al. Gliomas induce and exploit microglial MT1-MMP expression for tumor expansion. Proc Natl Acad Sci U S A 2009; 106 (30) 12530-12535
• 20 Ye XZ, Xu SL, Xin YH. et al. Tumor-associated microglia/macrophages enhance the invasion of glioma stem-like cells via TGF-β1 signaling pathway. J Immunol 2012; 189 (01) 444-453
• 21 Jimsheleishvili S, Alshareef AT, Papadimitriou K. et al. Extracranial glioblastoma in transplant recipients. J Cancer Res Clin Oncol 2014; 140 (05) 801-807
• 22 Müller C, Holtschmidt J, Auer M. et al. Hematogenous dissemination of glioblastoma multiforme. Sci Transl Med 2014; 6 (247) 247ra101
• 23 Nilsson RJ, Balaj L, Hulleman E. et al. Blood platelets contain tumor-derived RNA biomarkers. Blood 2011; 118 (13) 3680-3683
• 24 Meikle CK, Kelly CA, Garg P, Wuescher LM, Ali RA, Worth RG. Cancer and thrombosis: the platelet perspective. Front Cell Dev Biol 2017; 4: 147
• 25 Placke T, Kopp HG, Salih HR. Modulation of natural killer cell anti-tumor reactivity by platelets. J Innate Immun 2011; 3 (04) 374-382
• 26 Han S, Liu Y, Li Q, Li Z, Hou H, Wu A. Pre-treatment neutrophil-to-lymphocyte ratio is associated with neutrophil and T-cell infiltration and predicts clinical outcome in patients with glioblastoma. BMC Cancer 2015; 15: 617
• 27 Subeikshanan V, Dutt A, Basu D, Tejus MN, Maurya VP, Madhugiri VS. A prospective comparative clinical study of peripheral blood counts and indices in patients with primary brain tumors. J Postgrad Med 2016; 62 (02) 86-90
• 28 Daugherty SE, Pfeiffer RM, Sigurdson AJ. et al. Nonsteroidal antiinflammatory drugs and bladder cancer: a pooled analysis. Am J Epidemiol 2011; 173 (07) 721-730
• 29 Brockmann MA, Bender B, Plaxina E. et al. Differential effects of tumor-platelet interaction in vitro and in vivo in glioblastoma. J Neurooncol 2011; 105 (01) 45-56