Survival of Patients with Cancer Associated Thrombosis at the Uganda Cancer Institute

Survival of patients with cancer associated thrombosis at the Uganda Cancer Institute

Clement D Okello, Yusuf Mulumba, Abrahams Omoding, Henry Ddungu and Jackson Orem

Uganda Cancer Institute, Upper Mulago Hill Road, P.O. Box 3935, Kampala, Uganda

Abstract

Background: The occurrence of venous thromboembolism (VTE) in patients with cancer leads to a reduced life expectancy. There is an increased incidence of cancer and its associated mortality in Uganda. We described the survival and characteristics of patients with cancer associated thrombosis (CAT) in a tertiary oncology centre in Uganda.

Methods: We performed a retrospective study on patients with CAT at the Uganda Can- cer Institute (UCI) using a homogenous purposive sampling method.

Results: One hundred and eleven patients with documented VTE were included in the analysis. At entry, the mean age was 52.4 years, and 69 were female. Ninety eight had deep venous thrombosis, while 12 had pulmonary embolism. The most common cancer Research diagnoses were haematologic (30), gynaecologic (20) and prostate (17) cancers. Treat- ment regimens included anticoagulation with low-molecular weight heparin (LMWH) (72) and combined LMWH with warfarin (22). The median overall survival (OS) was 6.3 months, with a 1-year survival rate of 41.5%. Patients with significantly increased hazard of mortality were those with upper gastrointestinal (UGI) malignancies, colorectal and breast cancers. Patients with a body mass index of 25–29.9 kg/m2 (overweight) had a slightly reduced hazard of mortality.

Conclusion: The OS of patients with CAT at the UCI is short. Most patients with CAT presented with advanced stage cancers and at a relatively young age. Patients with UGI, Correspondence to: Clement D Okello colorectal and breast cancers had increased hazards of mortality, whereas those who Email: [email protected] were overweight had a slight reduction in the hazard of mortality. ecancer 2021, 15:1212 https://doi.org/10.3332/ecancer.2021.1212 Keywords: thrombosis, cancer, Uganda Published: 25/03/2021 Received: 03/11/2020 Publication costs for this article were supported by ecancer (UK Charity number 1176307). Background Copyright: © the authors; licensee ecancermedicalscience. This is an Open Access Patients with cancer have a 4- to 7-fold increased risk of suffering from venous thrombo- article distributed under the terms of the embolism (VTE) events, including deep venous thrombosis (DVT) and pulmonary embo- Creative Commons Attribution License (http:// creativecommons.org/licenses/by/3.0), which lism (PE) [1], a risk that is further augmented by cancer treatment [2, 3]. Development permits unrestricted use, distribution, and of VTE in the setting of cancer is associated with a significantly reduced life expectancy reproduction in any medium, provided the original [3, 4]. work is properly cited.

ecancer 2021, 15:1212; www.ecancer.org; DOI: https://doi.org/10.3332/ecancer.2021.1212 1 of patients with cancer-associated VTE inatertiary oncology centre inUganda. risk groupings ofpatients hospitalised [8]. Therefore, we undertook aretrospective study to thesurvival describe andclinicalcharacteristics have majorly been on case reports, clinicalexperience of radiological diagnosis and treatment, postmortem findings of deaths hospital and performedin studies sub-Saharanthe outside Africa mortality Analyses of theKampalacancer registry data show asteady increase intheincidence of cancers inUganda [5], with acorresponding risein VTE. Additionally, combination chemotherapy, includingantiangiogenic agents, alsoreportedly heightens therisk[3,4]. mary, andof theovary andbrain; patients locallywith advanced cancers andthose with distant metastases may alsobeat increased riskof a predictor, with thestrongest association being with malignant tumours of the pancreas, lung, stomach, adenocarcinomas of unknown pri- There are several factors that have been implicated in increasing the risk of thrombosis in patients with cancer. The type of cancer is a strong ecancer 2021, 15:1212; www.ecancer.org; DOI:https://doi.org/10.3332/ecancer.2021.1212 survival were illustratedthe Kaplan–Meier using curves. Survival was calculated from thedate of initialdiagnosis with thrombosis until the Demographiccharacteristicsand clinical were described usingfrequencies andpercentages. One year overall survival (OS) rate andmedian Data analysis Technology (UNCST). All patient information was anonymised. were obtained from theUganda Cancer Institute Research Ethics Committee (UCIREC) andtheUganda National Council for Science and (Epidata association, Denmark) before exportingto STATA Version 14(StataCorp, USA) for analysis. Study approvals and waivers of consent racy byinvestigatorprincipal the prior to acceptance for entry. Data were thencoded, andentered into a computer using Epidata version 3.1 Data were manually derived usingastandard data collectiontool. Completed data collection tool was checked for completeness andaccu- Data collection radiologically confirmed DVT, PEor other VTE inassociation ahistologicallywith confirmed cancer diagnosisat theUCI. ofCharts eligiblepatients were selecteda homogenous using purposive samplingmethod. Data were derived from chartsof patients with Eligibility criteria maintained inadesignated space at theUCIrecords offices. treated asoutpatients oralwith medication. Prescriptions are normally recorded on the patient medical records/charts, which are securely sis, medicationsthatand LMWH as heparin such require are injections given by nurses. thehospital Where possible,patients with CAT are Diagnosis of atVTE theUCIisundertaken usinginternationally recognised methodologies, primarily with vascular imaging[9].Upon diagno - weight heparin(LMWH),unfractionated heparin, warfarin andrivaroxaban. apy andsurgery. Patients CATwith are alsotreated atUCI. the Available treatment options for thrombosis at theUCIincludelow-molecular The UCIprovides both inpatient andoutpatient services. Available options for cancer treatment at theUCIincludechemotherapy, radiother Democratic Republic of Congo, South Sudan, Kenya, Tanzania, Rwanda andBurundi, are alsotreated at theUCI. Most patients cancerwith inUganda seekcancer care at theUCI.Occasionally, patients from theneighbouringcountries, includingthe This was aretrospective study conducted at theUganda Cancer Institute (UCI).UCIistheonly tertiary cancer treatment facility inUganda. Study designandsetting Methods [6]. Moreover, venousthrombosis hasbeenconsistently reported asasignificant causeof mortality incancer patients, especially [7]. The limited data oncancer associated thrombosis (CAT) inthesub-Saharan African 2 -

Research ecancer 2021, 15:1212; www.ecancer.org; DOI:https://doi.org/10.3332/ecancer.2021.1212 ence of in patientsVTE cancerwith isanindependentpredictor of poor survival [10,11] Thesurvival median of patients CATwith inour study of 6.3months was inline with moststudiesthat have demonstrated thatthe pres - Discussion 25–29.9 kg/m colorectalcancers CI, 1.95–205.9,p=0.01)or= 20,95% (HR breast cancer= 8.2,95%CI,1.12–59.88,p0.04).Patients (HR aBMIofwith 6 months prior to diagnosis. PatientsVTE increasedwith hazard of mortality =11.7,95%CI,30–59.13,p<0.01), hadUGImalignancies(HR mortality were age, sex, body massindexof (BMI), type cancer, stage of cancer, presence of comorbidities, and exposure to chemotherapy and prostate cancer, 37.3%(95%CI, 10.0–65.5). Factors that were analysed at both univariable and multivariable levels for predictors of cers, 44.77%(95%CI,23.6–63.9);others were gynaecological cancers, 42.4%(95%CI,17.3–65.8),breast cancer, 38.1%(95%CI,6.1–71.6), the longest 1-year OS were those 87.5% (95%CI,38.7–98.14),lungcancer,KS, with 50%(95%CI,5.78–84.49%) andhaematological can- 3.40–13.93). The 1-year OS was 41.5%(95%CI,29.7–52.9%),andthe2-year OS was CI,24.1–47.4%)(Figure 35.6%(95% Patients 1 ). with All patients enrolled in the study were followed up for survival analyses. The median OS for all patients CATwith was 6.3 months (95% CI, Survival (2%), andunfractionated Heparin, 1(1%). (22%) were prescribedcombined LMWH with warfarin. Other anticoagulation treatments were warfarin only therapy, 3(3%),rivaroxaban, 2 fore, data ondurationof anticoagulationtherapy wereMost missing. patients, 72(72%), were prescribedonly LMWH therapy, while 22 prescriptionon the for allthepatients,there was noinformation available to confirm that thepatients completed theentire course. There- The majority of patients, 100(90.1%), were prescribed anticoagulation therapy. Although theintended duration of treatment was indicated Treatmentof VTE teristics of thestudy population. 64 (67.7%)patientswhile had not received any form of cancer treatment prior to thediagnosisof VTE. toapplicable leukaemia. Thirty-eightpatients(34.2%) had beenexposed to chemotherapy within 6months prior to thediagnosisof VTE, reported.staging The systemTNM wasto used stage solidtumours while Ann Arbor system was usedto stage lymphomas;staging was not presented with advanced cancer stage. Only four patients (3.6%) had early stage disease (haematologic), and 20 patients (18.0%) had no haematologicalcancers, 30(27%),gynaecologicalcancers, 20(18%)andprostate cancer, 17(15.3%). The majority of patients, 87(78.4%), (84.7%) hadaDVT,and 3(2.7%)hadboth 12(10.8%)hadaPE DVT andPE. The mostcommonly represented cancer diagnoses were tion (SD)) age of thestudy population was 52.4(15.4) years. There were more female, thanmalepatients. 69(62.2%), Ninety four patients ofCharts 111patients who were seenfrom2003 to 2019metstudy eligibility and were includedintheanalysis. The mean(standard devia- Results ratios (HRs) and95%confidence intervals were generated. Statistical significance was set at p<0.05(two-sided). Patients who wereto lost follow up wereanalysisin the included and were censored onthelastrecorded date of review at theUCI.Hazard patient characteristicsatand OS univariable andmultivariable analyses starting with known factors associated with survival andthenothers. day of death, or untilthey were administratively censored. Cox proportional hazard model was usedto evaluate theassociation between those without VTE (36%). VTE without those the survivalwith of patients cancerwith without showedVTE amarkedly reduced 1-year survival inpatients with VTE (12%)compared with diagnosis ofthe thrombosis 2 (overweight) hadareduced hazard of mortality= 0.1,95%CI,0.02–0.95,p0.04)(Table (HR 2). [12]. Additionally, areportof alarge European database comparing thesurvival of patients with cancer and VTE , with almost all patientswith almost dying within 6months of Table 1shows thebaselinecharac 3 -

Research ecancer 2021, 15:1212; www.ecancer.org; DOI:https://doi.org/10.3332/ecancer.2021.1212 Table 1.Baseline characteristics. None Two or more modalities Surgery Radiotherapy Chemotherapy Cancer treatment 6months prior to VTE diagnosis, Not applicable Stage IV Stage III Stage II Stage I Cancer stage, n(%) Other Dual malignancy (UGI+prostate ) Colorectal Lung Upper gastrointestinal (UGI) Kaposi Sarcoma (KS) Breast Prostate Gynaecological Haematological Type of malignancy, n(%) Unknown Obese (>30) BMI (kg/m Other sites of thrombosis DVT andPE PE DVT Typeof VTE, Female sex, n(%) Age, mean(SD) years Overweight (25–29.9) Normal weight (18.5–24.9) Underweight (<18.5) 2 ), n(%) n (%) Factor n (%) 52.4 (15.4) 64 (67.7) 38 (34.2) 20 (18.0) 31 (28.0) 17 (15.3) 20 (18.0) 30 (27.0) 75 (67.6) 12 (10.8) 94 (84.7) 69 (62.2) 14 (12.6) 05 (4.5) 01 (0.9) 03 (2.7) 09 (8.1) 01 (0.9) 09 (8.1) 01 (1.0) 05 (4.5) 06 (5.4) 07 (6.3) 08 (7.2) 08 (7.2) 05 (4.5) 02 (1.8) 03 (2.7) 06 (5.4) 11 (9.9) 50 (45) Data 4

Research ecancer 2021, 15:1212; www.ecancer.org; DOI:https://doi.org/10.3332/ecancer.2021.1212 Figure 1.Overall survival of patients withCAT. Table 2.Predictors of mortality. Missing Late stage Early stage Cancer stage Unknown Obese (30andabove) Overweight (25–29.9) weight (18.5–24.9) Underweight (<18.5) BMI group Female Male Sex Age (years) Variable 1.2 (0.65-2.06) 0.4 (0.14–1.01) 0.5 (0.22–0.92) 0.6 (0.22–1.84) 1.5 (0.37–5.99) 0.3 (0.06–1.91) 0.8 (0.24–2.58) CHR 1 (0.98–1.02) a (95%CI) 1 1 1 p-value 0.05 0.03 0.41 0.58 0.22 0.69 0.61 0.96 2.3 (0.37–13.9) 0.6 (0.10–3.41) 0.6 (0.18–1.99) 1.2 (0.26–5.93) 0.1 (0.02–0.95) 0.6 (0.26–1.35) AHR 0.6 (0.18–2.2) 1 (0.97–1.02) a (95%CI) 1 1 1 p-value 0.38 0.55 0.40 0.78 0.04 0.46 0.84 0.22 5

Research ecancer 2021, 15:1212; www.ecancer.org; DOI:https://doi.org/10.3332/ecancer.2021.1212 a Table 2.Predictors of mortality. (Continued) Note: AHR, Adjusted hazard ratio; CHR,Crude hazard ratio No Gynaecological cancer Yes other Yes No Chemotherapy before VTE diagnosis Yes No KS Yes No Breast cancer Yes No Haematological cancer Yes No Prostate cancer Yes No Lung cancer Yes No UGI cancer Yes No Colorectal cancer Yes Yes cardiovascular No Comorbidities 2.7 (0.64–11.48) 1.2 (0.66–2.21) 0.8 (0.45–1.43) 0.2 (0.03–1.45) 1.2 (0.47–2.98) 0.6 (0.14–2.34) 3.1 (1.22–7.94) 1.3 (0.66–2.51) 0.9 (0.47–1.6) 0.9 (0.4–1.82) 1 (0.42–2.21) 1 1 1 1 1 1 1 1 1 1 0.54 0.92 0.45 0.11 0.73 0.64 0.69 0.43 0.02 0.18 0.46 8.2 (1.12–59.88) 2.6 (0.55–12.45) 11.7 (2.3–59.13) 20 (1.95–205.9) 1.9 (0.88–4.02) 0.7 (0.26–1.83) 0.7 (0.32–1.62) 0.1 (0.01–1.71) 1.5 (0.35–6.61) 0.9 (0.12–6.84) 3 (0.65–13.42) 1 1 1 1 1 1 1 1 1 1 <0.01 0.10 0.46 0.42 0.12 0.04 0.58 0.22 0.92 0.01 0.16 6

Research ecancer 2021, 15:1212; www.ecancer.org; DOI:https://doi.org/10.3332/ecancer.2021.1212 caution. CT scanshadto bepaidfor by thepatients out-of-pocket. The smallstudy samplesize alsosuggests that our resultshould betaken with selected patientpopulation who were ableto pay for diagnostic the workupfor the VTE since compression/Doppler ultra soundscansand diagnoses. Distributionof thesediagnoses would have provided aclearer picture of survival. It isalsopossiblethat our study hadauniquely absense of theduration of anticoagulation for thepatients. Additionally, haematological andgynaecological cancers were all conglomerate We have addeddata onsurvival inpatients CATwith inthesub-Saharan Africa. However, we acknowledge somemajor limitations including form of advanced-stage cancer [12,27]. in Europe and Americanhave consistently reported in patientsVTE with cancer of thepancreas, brain, liver, myeloma, multiple ovary andany However, thegroupingof cancers into haematological andgynaecological typesinour study may explain their over-representation. Studies similar finding was reported inastudy fromIbadan, Nigeria, where prostate cancer was themostcommon cancer associated with VTE [26]. The threecommon most cancers associated thrombosiswith inour study were haematological, gynaecological andprostate cancers. A with CAT to thepossibleuseof hormonereplacement therapy andhormonalcontraceptives insomefemales of reproductive age [25]. datadoes not provide reason forobservation, this havestudies some attributed thehigher representation of thefemale sex amongpatients Furthermore, the findingof more females thanmalespresenting with CAT isconsistent with a Europeanour study hospital [24].Although in Uganda the medianagewith of 16.7 years [23]. The younger age of presentation inour study population (mean age, 52 years) may beareflection of theoverall population age distribution overweight patients (BMI:25–28and28–<30) hadlower mortality risks[22]. risk of mortality [21].Similarly, inalarge study thata cohortincluded of 3,408menand women diagnosed colorectalwith cancer intheUS, any cancer andbreast cancer mortality inacohort of older Caucasian women, patients inthecategorywith aBMI of overweight hadalower better OSs thanpatients a normalBMI with example,cohortin asingle-institution of 7,765patients gastric cancer,with those who were overweight, mildly or moderately obese,had Theof finding aslightly reduced hazard of mortality inoverweightpatients isratheris consistent surprising,but with other studies.For have noimmediate explanation for theincreased hazard of mortality inpatients with breast cancer. al a significantpredictor of death 1 within year of cancer diagnosis[18].In arandomised trialonpatients with colorectal cancer by Mandalà et a poor OS.In aretrospective analysis of two large databases inCalifornia-US thatincluded 68,142patients with colorectal cancer, VTE was therapy.lation Tetzlaff observation, we canonly speculate that thismight bedueto thepotential increased bleedinginpatients with GImalignanciesonanticoagu- Patients with UGIand colorectal cancers had significantly higher hazard of mortality. Although our data do not provide reasons for this which speciﬁcpatient information andtherapeutic regimens were often unknown [13]. in cancerOS patients with treatedVTE usingextendedis longer LMWH thanthat reported from large registry andpopulation studiesin tion, where1-year the survival was only 12% Nevertheless, the1-yearin our OS study of 41.5% was higher thanreported inanother population basedstudy onalarge European- popula with KS tosize thesmallsample of bothcancer(8) andlung KS (5).Notwithstanding, arecent study inKenya reported excellent survival inpatients in patients haematologicalwith higherthe malignancies, 1-year inpatients OS with KSandlungcancers shouldbetaken with cautiondue and pancreaticlung cancers at29.4, 27.4,15.5,2.4and1.9months, respectively [13]. Although our resultshowed asimilarly high1-year OS tively. InNew the Zealand study, survival was longestfor haematological malignancy at 44.4months, followed by prostate, bowel, breast, Interestingly,in New somestudies ZealandhaveUS andthe reported longer mediansurvivals of 13.5[13]and16.7[14]months, respec [19], patients with had asignificantlyVTE increased riskof mortality even after adjustingfor age, diseasesite andtreatment schedule. We [15]. etal [16, 17]have reported thatpresencethe of VTE inpatients with advanced gastrooesophageal cancers leadsto [20]. In another study to determine theeffect of BMIontheriskof cardiovascular, all-cause, [10]. Mostpatients in our study were prescribed(72%). It LMWH has beensuggested that the The peakincidence of CAT intheUKpatients occurred earlier amongfemales thanmales. 7 -

Research ecancer 2021, 15:1212; www.ecancer.org; DOI:https://doi.org/10.3332/ecancer.2021.1212 and interpreted the data; JO: Interpreted thedata. All authors read andapproved thefinalmanuscript. DesignedCDO: thestudy, interpreteddata the and wrotemanuscript. the AO: Interpreted thedata; Interpreted HD; thedata; YM: Analysed Authors' contributions pretation, andmanuscript writing. necessarily reflectthose of theUCI. Thesource funding hadnodirect roles inthedesignof thisprotocol, data collection, analysis andinter This work wasby funded UCI. the Anyand conclusionsfindings opinions, expressed inthismaterial are those of theauthor(s) anddonot Funding All the authors have declared noconflicts of interest. Competing interests All data generated or analysed duringthisstudy are article. published includedinthis Availability of data andmaterials Not applicable. Consent for publication UNCST (Reference number:HS2412). Waiver of consentand study approvals were obtainedfrom UCIRECthe (Reference 15-2017)andthestudy number: was registered at the Ethics approval andconsent to participate Declarations boembolism Standard deviation; UCI,Uganda Cancer Institute; UCIREC, Uganda Cancer Institute Research andEthics Committee; VTE, Venous throm - CAT, Cancer associated thrombosis;DVT, Deep venousthrombosis; Low LMWH, molecular weight heparin;PE,Pulmonary embolism;SD, Abbreviations ate theduration andoutcome of anticoagulation therapies prospectively. hazards of mortality, whereas those who were overweight hadaslightly reduced hazard of mortality. Further studiesare suggested to evalu - and ata relatively younger age compared toin developed those countries. Patients with UGI, colorectal andbreastcancers hadincreased In conclusion,our study suggeststhat there ofOS isashort patients with CAT. Most of thesepatients present with advanced stage cancers Conclusion 8 -

Research 11. 10. ecancer 2021, 15:1212; www.ecancer.org; DOI:https://doi.org/10.3332/ecancer.2021.1212 References None. Disclaimers None. Previous presentation script. Kakemboand her teamfor retrievingand datacharts the abstraction. We alsothankDr Ed Conway for thethorough review of themanu- The authors thankallthepatients whose information were usedfor thisstudy. The authors alsothankallthestudy staffs especially Joweria Acknowledgments 9. 8. 7. 6. 5. 4. 3. 2. 1. common cancers Arch Intern Med 166(4)458–464https://doi.org/10.1001/archinte.166.4.458 PMID:16505267 Chew HK, Wun T, andHarvey D, etal (2006)Incidence of venous thromboembolism anditseffect onsurvival patients among with 343(25) 1846–1850https://doi.org/10.1056/NEJM200012213432504 SØrensen HT, Mellemkjaer L,andOlsen JH, PMID: of diagnosis embolism: venousthromboembolism Lim W, Le Galand Bates G, SM, 1(2) 87 Omunakweand Nwagha HE TU. Cancer-associatedthrombosis insome African populations: agrowing concern org/10.1182/asheducation-2016.1.196 Streiff(2016) Thrombosis MB in thesetting of cancer worldwide for 36cancers in185countries CA Cancer J Clin Bray F, Ferlay J, andSoerjomataram I,etcancer al (2018)Global statistics 2018:GLOBOCAN estimates of incidence andmortality 135(2) 432–439https://doi.org/10.1002/ijc.28661 PMID: Wabinga Nambooze HR, and S, AmulenPM, 28(151):180119 https://doi.org/10.1183/16000617.0119-2018 PMID:30918022 Fernandes CJ, Morinaga LTK,and Alves JL Jr, et (2019)Cancer-associatedal thrombosis: the when, how and why treatment Cancers 10(10)380https://doi.org/10.3390/cancers10100380 PMCID:6209883 Abdol Razak NB, Jones G, and Bhandari M, etal (2018) Thromb Hemost Hamzaand Mousa MS SA (2020)Cancer-associated thrombosis: riskfactors, molecular future mechanisms, management 26(1) e24–e40https://doi.org/10.1002/onco.13596 Streiff MB, AbutalibSA, andFarge D, et al (2021) 30482764 PMCID: 261076029620954282https://doi.org/10.1177/1076029620954282 PMID:32877229PMCID:7476343 6258916 etsociety al(2018)American of hematology 2018guidelinesfor management of venous thrombo- PMID: et al(2014)Trends in the incidence of cancer in Kampala, Uganda 1991–2010 et al(2000)Prognosis of cancers associated with venous thromboembolism PMCID: 6142498 27913480 PMCID: Update onguidelinesfor themanagement of cancer

Blood Adv Cancer-associated thrombosis: anoverview offactors, mechanisms,risk and 24615279 68(6)394–424https://doi.org/10.3322/caac.21492 PMID:30207593 Hematology Am Soc Hematol EducProgram 2(22) 3226–3256https://doi.org/10.1182/bloodadvances.2018024828 PMID:11117976 ‐associated thrombosis 2016(1) 196–205https://doi. Int J Noncommun Dis Eur Respir Rev N Engl J Med Int J CancerInt J Oncologist Clin Appl 9

Research 27. 26. 19. 18. 17. 16. 25. 24. 23. 22. 21. 20. 15. ecancer 2021, 15:1212; www.ecancer.org; DOI:https://doi.org/10.3332/ecancer.2021.1212 12. 14. 13. the generalpopulation: a population-based cohort study inDenmark, 1997–2006 Cronin-Fenton DP, Søndergaard F, andPedersen LA, Kotila T, Fasola F, andBusari E(2013) cancer patients:the ‘GISCAD-alternating schedule’study findings Mandalà M,Barni S,andFloriani I,et al (2009)Incidenceclinical implications and of venous thromboembolism inadvanced colorectal vival Alcalay A, Wun T, andKhatri V, etal (2006)Venous thromboembolism inpatients with colorectal cancer: incidence andeffect onsur org/10.1111/j.1442-2050.2008.00829.x PMID: radiotherapy for localizedcarcinoma of gastroesophagealand theesophagus junction Tetzlaff ED, Correa AM, andKomaki R,et (2008)Significanceal of thromboembolic occurring phenomena before andduringchemo - 1989–1995 protocolchemotherapy inpatients with advanced gastroesophageal adenocarcinoma Tetzlaff ED, Correa AM, andKomaki R,et (2007) alThe onsurvival impact of thromboembolic phenomenaoccurring before andduring https://doi.org/10.1186/s13027-019-0242-9 sj.bjc.6605883 PMID: Res Faiz AS, KhanI,andBeckmanMG, ciated thrombosispatientsamong at seen outpatient DVT clinics Paneesha S,McManus A, and Arya R, https://www.worldometers.info/world-population/uganda-population/ diagramsusing causal Kroenke CH,Neugebauer R,andMeyerhardt J, et al(2016)Analysis of body mass index and mortality in patients with colorectal cancer 17627012 of osteoporotic fractures Reeves KW, Faulkner K,andModugno F, et al(2007)Body massindex andmortality amongolder breast cancer survivors inthestudy 21(6) 913–924https://doi.org/10.1007/s10120-018-0818-x 29651648 PMID: and Joo J, B, LeePark JH, chemotherapycombination and antiretroviral therapy at Moiteaching andreferral hospital,Kenya BusakhalaKigenN, and G, Waako P, et al(2019)Three year survival patients among with aids-related Kaposi sarcoma treated with CMAR.S47094 versus those risk analysiswithout malignancy: usingMedicare dataclaims Levitan N,Dowlati A, Remick SC,et al(1999)Rates of andrecurrent initial thromboembolic diseaseamongpatients with malignancy Amer(2013) Cancer-associated MH thrombosis:presentationclinical andsurvival thrombus locationMed Intern J42(1)71–74https://doi.org/10.1111/j.1445-5994.2010.02401.x Prestidge T, Leeand Harper S, P, et al(2012)Survival inpatients with malignancy and venous thromboembolism by tumour subtype and 199909000-00001 PMID: 136(3)535–541https://doi.org/10.1016/j.thromres.2015.06.036 PMID: JClinOncol 24(7)1112–1118https://doi.org/10.1200/JCO.2005.04.2150 PMID: PMID:

JAMA Oncol 20842120 PMCID: 23926439 PMCID: et al(2018)Bodyindex mass mortality and inpatients with gastric cancer: alarge cohort study Cancer EpidemiolBiomarkers Prev 10499070 2(9) 1137–1145https://doi.org/10.1001/jamaoncol.2016.0732 PMID:27196302PMCID:5016213 et al(2015)Characteristicsfactors andrisk of cancer associated venous thromboembolism etal (2010)Frequency, demographics(according andrisk to tumour typeor site) of cancer-asso- A revisit of venous thromboembolism Afr J Med Med Sci 42(2)177–181 2965880 3732200 18459989 etal (2010)Hospitalisation for venous thromboembolism incancer patients and 16(7) 1468–1473https://doi.org/10.1158/1055-9965.EPI-07-0051

Thromb Haemost EurCancer J Medicine 26168693 PMCID: 45(1) 65–73 BrCancer J 103(2) 338https://doi.org/10.1160/TH09-06-0397 78(5) 285–291https://doi.org/10.1097/00005792- Cancer Manag Res 16505431 Cancer Interdiscipl Int J Am Cancer Soc https://doi.org/10.1016/j.ejca.2008.09.005 Dis Esophagus 103(7) 947–953https://doi.org/10.1038/ 4643665 Infect Agent andCancer 5 165https://doi.org/10.2147/ 21(7) 575–581https://doi. Gastric Cancer 14(1) 1–8 109(10) Thromb PMID: 10 -

Research