Chemotherapy induced thrombocytopenia treated by four types of platelets concentrates

Chemotherapy induced thrombocytopenia treated by four types of platelets concentrates A Ljubinka I. Nikolić1, Ninoslav D. Nedeljković2, Svetislav B. Jelić2, Nada D. Suvajdžić Vuković3, Ivana M. Fillipović-Lješković2, Srdjan Z. Marković4, Drina Lj. Janković5, Dragana A. Kastratović4 A 1 Clinic for Gynecology and Obstetrics, Clinical Centre of Serbia, Belgrade, Serbia 2 Institute for Oncology and Radiology of Serbia, Belgrade, Serbia 3 Clinic for Hemathology, Clinical Centre of Serbia, Belgrade, Serbia 4 Clinical Centre of Serbia, Belgrade, Serbia 5 Vinča Institute of Nuclear Sciences, Belgrade, Serbia


INTRODUCTION
Serious adverse event of anti cancer chemotherapy is glanulocytopenia and thrombocytopenia which can decrease effi ciency of fi nal therapy results.
Th e main complications of granulocytopenia is neutropenic sepsis, usually designed as febrile neutropenia.Th e consequences of leukopenia can be prevented in part of hospitalization in bacteriologically protected units and sterile box and profi lactic usage of granulocate-collony stimulatinf factor (G-CSF) and granulocyte-macrophage colony stimalating factor (GM-CSF).Treatment includes a lot of available G-CSF and the CSFs.
Beside there are no such preventive measures and treatment facilities for thrombocytopenia which can result in life-threatening haemorrhagic syndrome, and platelet transfusion remains the only eff ective approach.Although both autologous bone marrow transplantation and periferal blood stem rescue have been advocated for high dose chemotherapy haematopoietic support, there is no doubt for importance platelet transfusions in the management of chemotherapy-related haematological toxicities in cancer patients.Advances in platelet transfusion have contributed to improved outcomes in the treatment of cancer patients, allowing high dose chemotherapy delivery with increased safety [1,2].
Continuous supportive platelet therapy is enabled by technological advancement and increased quality of plastic blood bags adjusted for prolonged platelet storage.However, concentrate-specifi c parameters can be infl uenced by preparation technique and storage procedures.Th e duration of storage causes increase of TGF level, which induces an impressive decrease in the in-vivo platelet recovery and survive [3,4,5,6].
Th e benefi ts of prophylactic platelet transfusion are controversial and the circulating platelet level which predisposes to haemorrhage is unsertain; so is the eff ectiveness of alternatives to platelet therapy.Finally, the merits and drawbacks of treatment of thrombocytopenic patients with platelets from single donors and multiple random donors, need evaluations [7,8].
Several serious scientists through the world aft er many years are still researching problem of platelet transfusion, but there is no for sure about standpoint [9,10,11].
It has been shown by necessity that many patients can tolerate prolonged periods profound thrombocytopenia without serious problems.A two-fold (double) prolongation of the normal bleeding time, has been reported to be indicative of signifi cant risk of bleeding [12].

THE AIM
To determine whether there is a diff erence in the clinical effi ciency in the use of 4 types of platelet applied for transfusion; -to ascertain whether platelet count increase expressed as corrected count increment (CCI), is a better parameter for the evaluation of platelet transfusion effi ciency than the bleeding time (Bt), as the only readily assessable in vivo platelet function related parameter.

SUBJECTS AND METHODS
Th is paper is a part of academic (noncommercial) IV phase observational nointervetion study, done at National Institute of Oncology and Radiology of Serbia (NIORS), Belgrade.All patients signed informed consent.
Because of technical problems this study was frozen, and this year problem was recognized again so these results are the fi rst step of continuing ivestigating platelets charactreristics.
Clinical bleeding during chemotherapy-induced thrombocytopenia is a serious clinical problem.During several decades this serious citostatic adverse event is not solved and there is no doctrinar statement about this.
Cancer as itself disease is characterised with non stabile hemostatic parameters [13], so it is case to case clinical situation.
Investigation included 78 patients (34 males and 44 females), median age 57 years (range 23-73), diagnosed with limphoma and metastatic solid tumors, transfused by platelet concentrates from January till May 1998.None of the patients had disseminated intravascular coagulation (DIC), wide spread adenocarcinoma, infection, or other special situation or diagnosuis that might have infl uenced the outcome of platelet support.
Th e standards for research laboratories procedures, and investigator were met the required criteria.
Patients were devided into 4 groups, based on the type of platelet concentrates used for transfusion, and produced in Serbian National Blood Transfusion Institute (SNBTI).Four types of platelet concentates were used depending on Institute supply by the producer.Group A -patients transfused with leucode-pleted platelet concentrates (PC) pool obtained from 5 platelet rich plasma PC units (PRP PC) ABO identical, or from 5 buff y coat PC units (BC PC) ABO identical (Leukotrap pooling system®, Cutter USA) Gropu B -patients transfused with prestorage fi ltered PRP PC or BCPC using fourth generation fi lters (Terumo, Japan) Group C -patients transfused with standard non-leukodepleted PC (PRP-PC stored 0-5 days) (plastic bags Terumo, Japan) on horizontal shaker at 22°C Group D -patients transfused with prestorage leukodepleted PC obtained from random donors BC, using spontaneus sedimentation of verticaly positioned and stored BCs for 12 hours at 22°C.In group A there were 3 patients whith limphoma and 24 with metastatic solid tumors; In group B there was one patients with limphoma and 6 patients with metastatic solid tumors; In group C there were 5 patients with limphoma and 28 with metastatic solid tumors and , group D there were 11 patients with metastatic solid tumors.Th us, there were 9 patients with lymphoma and 69 with metastatic solid tumors.Th rombocytopenia in all those patients was serious adverse event (SAE), as consequence of needed aggressive chemotherapy.
Indication for platelet transfusion treatment were: 1. bleeding time prolonged in 108/129 transfusion (Bt 3.4 min -37 min); 2. when clinical manifestation of haemorrhagic syndrome there present in 2/129 transfusions; 3. when bleeding time was normal but platelets already with WHO grade 4 toxicity range, tested daily expressed tendency towards decrease in 19/129 transfusions.
Out of 108 transfi sions in patients with prolonged bleeding time, thrombocytopenia was grade 4 in 97 patients, grade 3 in 8 cases, grade 2 in 2 cases and grade 1 in 1 case.
None of patients had DIC.Grading of thrombocytopenia was performed according to CTC criteria [14].
Effi ciency of supportive therapy using PC transfusions was estimated according both to CCI and Duke test for bleeding time (BT) (reference values1 -3 minutes), not traumatic oobtained from blood specimens using the EDTA.Platelet values were repeated using the sodium citrate and the counting was carried out microscopically in chamber with the 1% ammonium oxalate.Platelet count was not tested in our patients 1 h aft er PC transfusion because in that interval only data concernic total number of transfused platelet particles can be obtained.Storage platelets loose viability, decrease 2,3diphosphoglycerate (2,3 DPG), but can be acceptable as compromise during storage [16].Around 20-24 hours aft er transfusion, removal of damaged platelets occurs, and energetic balance of platelets which have lost their energetic balance during storage is restored.
CCI was determined using the folloing formula: CCI = ((post transfusion platelet count (x10 9 / L)−pretranfusion platelet count (m 2 ))/total number platelets transfused It is generally accepted that an eff ective PC treatment corresponds to a CCI aft er 24 h which is ≥ 4.5.; for the purpose of this study we have considered as eff ective any CCI over zero (>0) i.e. any platelet increase its relation to pretransfusion one.A corected bleeding time was considered any shorthening below 6 minutes if initial bleeding time was over 6 minutes; if initial bleeding time was 6 minutes or below any shorthening was considered as evidence for a corected bleeding time, meaning that clinical endpoints (i.e.bleedeing) are the most important method for evaluating effectiveness of PLT transfusions [17].
In addition on PC investigated patients were transfuzed with red blood cells (RBC) (concentrates), fresh frozen plasma (FFP) and cryoprecipitate (Cryo): in group A were transfused with RBC and 2 with FFP; in group B , 2 patients were transfused with RBC; in group C, 10 patients were transused with RBC; One of those 10 patients also receive cryo and another received RBC and FFP.In group D two patients RBC.
Statistical methods used in the investigation were Kruskal-Wallis, Mann Whitny and Wilcoxon Rank Sum W test.

RESULTS
Th e patient's characteristics are presented on table 1.Out of 78 patients, 33 (41%) were transfused with non-leukodepleted PC, and remaining 45 (59%) with leukocyte depleted PC.Total number of 647 PC units were used.Out of 647 PC units, 235 units (36.3%) were non-leukodepleted and 412 (63.7%) units were leukodepleted.Patients were transfused with the total number of 129 PC transfusions.Mean number of PC transfusions per patient was 8.3 PC units, and 4.8 PC unit per one transfusion (table 1).

DISCUSSION
Th e investigated 4 groups were homogenous before platelet transfusions according to platelet count (p=0.4) and not homogenous according to Bt (p<0.01 tj p=0.006).Aft er supportive therapy using PCs Bt was corrected and became similar in all 4 goups.Nonhomogenity among 4 groups dessappeared (p=0.44) which possibly pointed to the positive eff ects of supportive platelet therapy and successful therapeutic result.Th ese results correllated with literature data [3,7] Concerning platelet count and Bt of all investgated patient groups there were no signifi cant correlation either before (ρ= -0.08) ar aft er PC transfusions (ρ= -0.16) [9].
Signifi cant platelet increase was noted in group A (p<0.05 tj p=0.046) and group C (p=0.0001), in patients transfused with leukodeplated platelets (groups A+B+D) (p<0.05 p=0.0239), as well as in all patients together (groups A+B+C+D) (p<0.00001).In groups B and D the number of patients was small so that signifi cance could not be determine, although the increase was noted.
Compared with group A, group C had signifi cantly higher platelet count aft er PC transfusion (p=0.0048), as well as a higher CCI compared with group A (p=0.0248).According to post transfusion Bt, groups C and A did not diff er signifi cantly (p=0.9694).
Groups threated with leukocyte depleted PC (groups A, B and D) and group threated with non-leukocyte depleted PC showed homogenity before transfusion both according to platelet count (p=0.8) an Bt (p=0.07).
Non-leukodepleted PC treated group (group C) compared with leukodepleted PC treated group (A+B+D) showed signifi cantly higher platelet count increase (p<0.01 tj.p=0.0082), and signifi cantly higher CCI (p<0.05p=0.0139)aft er PC transfusion.Higher effi ciency of non-leukodepleted PC could be explaned partly by the presence of plasma and leukocyte contents (PGE1, IL6) [18,19,20,10], and partly by the loss of a population of platelets during leukodepletion procedure (in groups A, B, D) and the loss of cytokines which take part in platelet function and viability maintenance.However nonleukodepleted platelets are not recomended if multiple plartelet transfusion are required, in order to decrease the exposure to alloantigens, i.e. in order to postpone alloimmunization and refractorines [21,22,23,24,11].
Despite of non-uniforme numeric results and wide diferences in types of malignat diseases, clinicaly transfusion therapy outcome was satisfactory which is in correlation literatures dates.Aft er more than ten years there is still dillemas and litlle bit confusion rellated to recomended values of platelet increase, shortening bleeding time and clinical outcome [9,13,21].
Bt aft er PC transfusiomn showed no statistical diff erece between the groups of leukodepleted and non-leukodepleted PC treated patients (p=0.61).Leukocyte presence in group C did not induce occurence of posttransfusion reaktions.One posttransfusion reaction occured in group A and could be explained by the reaction to plasma present in platelet supernatant [21].
BT were signifi cantly corrected in all patient groups in relation with the BT prior to platelet transfusion(p<0.01).
In 23 PC transfusions platelet count and CCI increase did not occur.Out of 23 PC transfusions with unsatisfactory CCI, 5 were performed in 5 males, and 18 in 13 female patients.None of the female patents was treated by any other blood product.Likewise, the only posttransfusion reaction occurred in a female patient.As all there females were uni or multiparous this could speculatively be explained by the contact with another antigenic system during pregnancy [25,26].
In 20 out of 23 PC transfusions where platelet increase did not occurre and satisfctory CCI was not achieved, BT was corrected.In all 4 investigated groups BT was corrected in signifi cantly higher percentage of PC transfusions compared with CCI (p=0.002).
To improve effi ciency of platelet transfusion in patients who develope refractoriness, and to prevent or delay the onset of refractoriness, the storage time should be shortened [3,27,11]; single donor platelets (SD PC) should be used whenever possible rather than standard PC [28,11].Transfusion of high platelet doses increases the transfusion inter-val and can reduced the number of Platelet concentrates required by thrombocytopenic patients that signifi cantly reducing donor exposure [29,11].HLA an HPA-matched platelets [30,31]; or as an alternative HLA class I eluted platelets should be used for succesful treatment of platelet transfusion refractoriness [32].Washed platelet concentrates in patients with febrile non-haemolytic transfusion reactions should be used [33].Autologous platelet transfusions are the best choice [34,11].
Disadvantages using profi lactic PC therapy include increased number of donor exposures and possibly also enhanced risk of alloimmunzation and refractorines [35,36,10,11].Th e advantages of profi lactic therapy are still not universaly accepted.
Even when the value of prophylactic therapy is accepted the platelet count threshold at which this is justifi ed is still very much a matter of debate and it is certainly true that not all patients with severe thrombocytopenia should be considered autothomatic candidates for prophylactic therapy [37,38,39].
Since, many patients can tolerate prolonged periods of profound thrombocytopenia without serious bleeding problems [10,11], bleeding time is much better parameter to assess (for assessement) the need for platelet transfusions than platelet level.
Platelet transfusion practice is being questioned more than ever before.As we develop better therapies and guidelines, the practice of platelet therapy can be expected to change in the near future [10,11].

CONCLUSION
Th e advantages of profi lactic PC therapy are not still universaly accepted, but authors agree with other published papers, that prophilactic platelet transfusion is clinicaly recommended.
Advantages using profi lactic PC therapy include expenses preventing further costs of bleeding complications.Disadvantages are increased number of donor exposures and possibly also enhanced risk of alloimmunization.
Th ese data show that beside PC therapy effi ciency monitoring, which includes platelet counts and CCI, Bt in vivo platelet function test should also be performed.Since bleeding time was corrected in cases of uncorrected CCI, uncorrected CCI should not be considered as refractorines to PC without the deter-mination of bleeding time.PC transfusions followed by a satisfactory CCI but uncorrected Bt associated with clinical imrovement failed should not be considered as successful ones.
In all 4 investigated groups of patients bleeding time was a far better parameter compared with CCI for the PC therapy effi cency .Bleeding time is much better parameter than platelet level for assessement the need for platelet transfusions.
Considering to cancer disease ethiology, course of disease, and patient clinical status in this sofi sticated part of treatment success, it is necessery to assese clinical cost benefi t results in any patient separetly.
Authors sugest to be carrefull and follow clinical and laboratory results personalised to single patient.
Th ere is a need to develope better therapies and guidelines so the practice of platelet therapy can be expected to improve in the future.In the next part of this investigation we shall underline personalized medical treatment.

Table 2 .
Indication for platelet transfusion

Table 3 .
Results are showen in table 7. Platelet count was noted in a signifi cantly lesser Main www.hophonline.org

Table 4 .
Platelet counts before and after PC transfusion

Table 5 .
Bleeding time before and after platelet transfusion

Table 7 .
Analysis of Platelet transfusion Effi ciency In 4 PC transfusions (4 patients) was not corrected 20-24 hors aft er the procedure, while platelet count had increased.CCI was satisfactory in 47/52 (92.1%) and Bt was corrected in 49/51 (96.1%)PC transfusions.According to CCI and/or Bt , PC transfusions gave satisfactory result in 51/51 (100%) transfusions in group C. Bt before PC transfusion was 9.9±6.3min,median=8.8,range 3.4-37 minutes.Bt was not corrected in one female patients with NHL and in one patient with testicular cancer.CCI was not satisfactory in the patient with NHL, who had increase platelet count and uncorrected Bt in previous PC * No of patients/no of plt transfusions, BT-bleeding time, PC-platelet concentrate received any other blood products.Based on CCI, 8/10 PC transfusions, and according to Bt, all showed satisfactory response.Group C: transfusion, as well as in the patient with metastatic solid tumors (leiomyosarcoma of the uterus, in one with gastric cancer, and in one with ovarian cancer).In this group there were no immediate transfusions reactiones (tabl 7).