Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review)

Akl EA, Gunukula S, Barba M, Yosuico VED, van Doormaal FF, Kuipers S, Middeldorp S, Dickinson HO, Bryant A, Schünemann H

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2011, Issue 4 http://www.thecochranelibrary.com

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. TABLE OF CONTENTS HEADER...... 1 ABSTRACT ...... 1 PLAINLANGUAGESUMMARY ...... 2 SUMMARY OF FINDINGS FOR THE MAIN COMPARISON ...... 2 BACKGROUND ...... 5 OBJECTIVES ...... 7 METHODS ...... 7 RESULTS...... 9 Figure1...... 12 Figure2...... 13 Figure3...... 14 Figure4...... 15 Figure5...... 17 Figure6...... 19 Figure7...... 20 DISCUSSION ...... 20 AUTHORS’CONCLUSIONS ...... 21 ACKNOWLEDGEMENTS ...... 22 REFERENCES ...... 22 CHARACTERISTICSOFSTUDIES ...... 26 DATAANDANALYSES...... 48 Analysis 1.1. Comparison 1 Heparin vs placebo, Outcome 1 Mortality at 12 months...... 48 Analysis 1.2. Comparison 1 Heparin vs placebo, Outcome 2 Mortality at 24 months...... 49 Analysis 1.3. Comparison 1 Heparin vs placebo, Outcome 3 Mortality over duration of study...... 50 Analysis 1.4. Comparison 1 Heparin vs placebo, Outcome 4 SymptomaticVTE...... 51 Analysis 1.5. Comparison 1 Heparin vs placebo, Outcome 5 Major bleeding...... 52 Analysis 1.6. Comparison 1 Heparin vs placebo, Outcome 6 Minor bleeding...... 53 APPENDICES ...... 53 WHAT’SNEW...... 55 HISTORY...... 56 CONTRIBUTIONSOFAUTHORS ...... 56 DECLARATIONSOFINTEREST ...... 56 SOURCESOFSUPPORT ...... 57 DIFFERENCES BETWEEN PROTOCOL AND REVIEW ...... 57 INDEXTERMS ...... 57

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) i Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. [Intervention Review] Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation

Elie A Akl1, Sameer Gunukula1, Maddalena Barba2, Victor E D Yosuico1, Frederiek F van Doormaal3, Saskia Kuipers3, Saskia Middeldorp3, Heather O Dickinson4, Andrew Bryant4, Holger Schünemann5

1Department of Medicine, State University of New York at Buffalo, Buffalo, NY, USA. 2Department of Epidemiology, National Cancer Institute Regina Elena, Rome, Italy. 3Department of Vascular Medicine, Academic Medical Center, Amsterdam, Netherlands. 4Institute of Health and Society, Newcastle University, Newcastle upon Tyne, UK. 5Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Canada

Contact address: Elie A Akl, Department of Medicine, State University of New York at Buffalo, ECMC CC-142, 462 Grider Street, Buffalo, NY, 14215, USA. [email protected].

Editorial group: Cochrane Gynaecological Cancer Group. Publication status and date: New search for studies and content updated (no change to conclusions), published in Issue 4, 2011. Review content assessed as up-to-date: 5 December 2010.

Citation: Akl EA, Gunukula S, Barba M, Yosuico VED, van Doormaal FF, Kuipers S, Middeldorp S, Dickinson HO, Bryant A, Schünemann H. Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagu- lation. Cochrane Database of Systematic Reviews 2011, Issue 4. Art. No.: CD006652. DOI: 10.1002/14651858.CD006652.pub3.

Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

ABSTRACT

Background

Anticoagulation may improve survival in patients with cancer through an antitumor effect in addition to the perceived antithrombotic effect.

Objectives

To evaluate the efficacy and safety of parenteral anticoagulants in patients with cancer with no therapeutic or prophylactic indication for anticoagulation.

Search strategy

A comprehensive search included (1) an electronic search (February 2010) of the following databases: Cochrane Central Register of Controlled Trials (CENTRAL) Issue 1, 2010, MEDLINE, EMBASE and ISI the Web of Science; (2) handsearch of conference proceedings; (3) checking of references of included studies; and (4) use of the ’related citation’ feature in PubMed.

Selection criteria

Randomized controlled trials (RCTs) assessing the benefits and harms of parenteral anticoagulation in patients with cancer but no therapeutic or prophylactic indication for anticoagulation.

Data collection and analysis

Using a standardized form we extracted in duplicate data on methodological quality, participants, interventions and outcomes of interest including all-cause mortality, symptomatic thromboembolism, major bleeding, minor bleeding and quality of life (QoL).

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 1 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Main results Of 8187 identified citations, nine RCTs enrolling 2857 patients fulfilled the inclusion criteria. In all included RCTs the intervention consisted of heparin (either unfractionated heparin or low molecular weight heparin). Overall, the effect of heparin therapy on mortality was not statistically significant at 12 months (risk ratio (RR) 0.93; 95% CI 0.85 to 1.02) but it was statistically significant at 24 months (RR 0.92; 95% CI 0.88 to 0.97). Heparin therapy was associated with a statistically and clinically important reduction in venous thromboembolism (RR 0.55; 95% CI 0.37 to 0.82). There were no statistically significant effects on major bleeding (RR 1.30; 95% CI 0.59 to 2.88), minor bleeding (RR 1.05; 95% 0.75 to 1.46) or QoL. The quality of evidence was high for symptomatic venous thromboembolism, moderate for mortality, major bleeding and minor bleeding, and low for QoL. Authors’ conclusions Heparin was associated with a significant reduction of death at 24 months but not 12 months. It was also associated with a reduction in venous thromboembolism but based on the RCTs in this review it had no significant effect on major bleeding, minor bleeding or QoL. Future research should further investigate the survival benefit of different types of anticoagulants in patients with different types and stages of cancer. The decision for a patient with cancer to start heparin therapy for survival benefit should balance the benefits and downsides and integrate the patient’s values and preferences.

PLAIN LANGUAGE SUMMARY Injectable blood thinners (anticoagulants) in patients with cancer Research evidence suggests that blood thinners may improve the survival of patients with cancer. This benefit could be related to a direct antitumor effect in addition to preventing blood clots. In this systematic review, data from nine trials and 2857 participants suggest that injectable blood thinners reduce the risk of blood clots by about half and possibly increase the risk of major bleeding by 30%, but the effect on survival remains unclear. The main limitations of this systematic review are the relatively small number of included trials and the inclusion of different types and stages of cancer.

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 2 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. aetrlatcauaini ainswt acrwohav who cancer with patients in anticoagulation Parenteral oyih 01TeCcrn olbrto.Pbihdb J by Published Collaboration. Cochrane The 2011 © Copyright SUMMARYOFFINDINGSFORTHEMAINCOMPARISON [Explanation]

heparin compared to no heparin for patients with cancer who have no other therapeutic or prophylactic indication for anticoagulation

Patient or population: patients with cancer who have no other therapeutic or prophylactic indication for anticoagulation Settings: outpatient Intervention: heparin Comparison: no heparin

Outcomes Illustrative comparative risks* (95% CI) Relative effect No of Participants Quality of the evidence Comments (95% CI) (studies) (GRADE)

Assumed risk Corresponding risk

oteaetco rpyatcidcto o anticoagul for indication prophylactic or therapeutic no e no heparin heparin h ie os Ltd. Sons, & Wiley ohn Mortality Medium risk population RR 0.93 2531 ⊕⊕⊕ Follow-up: 12 months (0.85 to 1.02) (8 studies) moderate1,2,3 649 per 1000 604 per 1000 (552 to 662)

Symptomatic VTE Medium risk population RR 0.55 2264 ⊕⊕⊕⊕ Follow-up: 12 months (0.37 to 0.82) (7 studies) high1 29 per 1000 16 per 1000 (11 to 24)

Major bleeding Medium risk population RR 1.3 2843 ⊕⊕⊕ Follow-up: 12 months (0.59 to 2.88) (9 studies) moderate1,4 7 per 1000 9 per 1000 (4 to 20) to (Review) ation Minor bleeding Medium risk population RR 1.05 2345 ⊕⊕⊕ Follow-up: 12 weeks (0.75 to 1.46) (7 studies) moderate1,4 27 per 1000 28 per 1000 (20 to 39) 3 aetrlatcauaini ainswt acrwohav who cancer with patients in anticoagulation Parenteral oyih 01TeCcrn olbrto.Pbihdb J by Published Collaboration. Cochrane The 2011 © Copyright

Health related quality of See comment See comment Not estimable5 0 ⊕⊕ life (1 study) low6 the Uniscale and the Symptom Distress Scale (SDS); Better indicated by lower values Follow-up: 12 months

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio;

GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. oteaetco rpyatcidcto o anticoagul for indication prophylactic or therapeutic no e

h ie os Ltd. Sons, & Wiley ohn Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

1 Vast majority of studies had allocation concealment , and used blinded outcome and adjudication. We did not downgrade although there was some concern about lack of blinding in some studies; the overall risk of bias was felt to be very low. 2 There is moderate heterogeneity among studies included in the analysis of death at 12 months (I2=41%). The subgroup analysis for mortality at 12 months was statistically significant and suggested survival benefit in patients with SCLC but not in patients with advanced cancer. Overall we decided to downgrade by one level when considering these issues along with imprecision. 3 CI interval includes effects suggesting benefit as well as no benefit. 4 CI includes possibility of both harms or benefits. 5 The scores for the 2 scales were similar for the 2 study groups, both at baseline and at follow-up 6 High risk of bias and only 138 patients enrolled. to (Review) ation 4 BACKGROUND Please refer to the glossary for the definitions of technical terms (Table 1).

Table 1. Glossary

Term Definition

Adjuvant therapy Assisting in the amelioration, or cure of disease

Anticoagulation The process of hindering the clotting of blood especially by treatment with an anticoagulant

Antithrombotic Used against or tending to prevent thrombosis (clotting)

Bacteremia The presence of bacteria in the blood.

Central venous line Synthetic tube that is inserted into a central (large) vein of a patient to provide temporary intravenous access for the administration of fluid, medication or nutrients

Coagulation Clotting

Deep vein thrombosis (DVT) A condition marked by the formation of a thrombus within a deep vein (as of the leg or pelvis) that may be asymptomatic or be accompanied by symptoms (as swelling and pain) and that is potentially life-threatening if dislodgment of the thrombus results in pulmonary embolism

Fibrin A white insoluble fibrous protein formed from fibrinogen by the action of thrombin especially in the clotting of blood

Fondaparinux An anticoagulant medication

Hemostatic system The system that shortens the clotting time of blood and stops bleeding

Heparin An enzyme occurring especially in the liver and lungs that prolongs the clotting time of blood by preventing the formation of fibrin. Two forms of heparin that are used as anticoagulant medications are: unfractionated heparin (UFH) and low molecular weight heparins (LMWH)

Impedance plethysmography A technique that measures the change in blood volume (venous blood volume as well as the pulsation of the arteries) for a specific body segment

Kappa statistics A measure of degree of non-random agreement between observers and/or measurements of a specific categorical variable

Metastasis The spread of a cancer cells from the initial or primary site of disease to another part of the body

Oncogene A gene having the potential to cause a normal cell to become cancerous

Osteoporosis A condition that affects especially older women and is characterized by decrease in bone mass with decreased density and enlargement of bone spaces producing porosity and brittleness

Parenteral nutrition The practice of feeding a patient intravenously, circumventing the gut

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 5 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Table 1. Glossary (Continued)

Pulmonary embolism (PE) Embolism of a pulmonary artery or one of its branches that is produced by foreign matter and most often a blood clot originating in a vein of the leg or pelvis and that is marked by labored breathing, chest pain, fainting, rapid heart rate, cyanosis, shock, and sometimes death

Stroma The supporting framework of an organ typically consisting of connective tissue

Thrombin A proteolytic enzyme formed from prothrombin that facilitates the clotting of blood by catalyzing conversion of fibrinogen to fibrin

Thrombocytopenia Persistent decrease in the number of blood platelets that is often associated with hemorrhagic con- ditions

Thrombosis The formation or presence of a blood clot within a blood vessel

Vitamin K antagonists Anticoagulant medications that are used for anticoagulation. Warfarin is a vitamin K antagonist

Warfarin An anticoagulant medication that is a vitamin K antagonist that is used for anticoagulation

Ximelagatran: An anticoagulant medication

Description of the condition Researchers have hypothesized that heparin may improve out- Since the 1930s, basic scientists have been exploring the effects of comes in cancer patients through an antitumor effect in addition anticoagulation on cancer (Smorenburg 2001). Studies have im- to its antithrombotic effect (Thodiyil 2002). In a 1992 clinical trial plicated the tumor-mediated activation of the hemostatic system comparing nadroparin, a low molecular weight heparin (LMWH), in both the formation of tumor stroma and in tumor metastasis to unfractionated heparin (UFH) in patients with proven deep (Dvorak 1986; Francis 1998; Levine 2003). There is also evidence vein thrombosis (DVT), nadroparin unexpectedly reduced mor- that heparin inhibits expression of oncogenes and formation of tality in the subgroup of patients with cancer (Prandoni 1992). thrombin and fibrin induced by cancer cells (Smorenburg 2001). At the same time, anticoagulants increase the risk for bleeding. In addition, heparin potentially inhibits intravascular arrest of can- This risk is higher in cancer patients compared to those without cer cells and thus metastasis (Smorenburg 2001). cancer. In fact, in patients with venous thrombosis on anticoagu- lation the risk of bleeding was higher if patients had cancer and correlated with the extent of cancer (Prandoni 2002). Heparins Description of the intervention are also known to cause thrombocytopenia and heparin-induced Heparin, low molecular weight heparins (LMWHs), fondaparinux thrombocytopenia (HIT) syndrome (Girolami 2006). and danaparoid do not have intrinsic anticoagulant activity but potentiate the activity of antithrombin III in inhibiting activated coagulation factors. These agents constitute indirect anticoagu- lants as their activity is mediated by plasma cofactors. Recombi- Why it is important to do this review nant hirudin, bivalirudin and argatroban directly inhibit thrombin In 1999, Smorenburg et al conducted a systematic review of the ef- and are classified as direct anticoagulants (Hirsh 2008). Heparin fects of UFH on survival in patients with malignancy (Smorenburg and its low molecular weight derivatives are not absorbed orally 1999). They found three trials of high methodological quality but and must be administered parenterally by intravenous infusion or with conflicting results. Since 1999 reports of several random- subcutaneous injections (Hirsh 1993). ized controlled trials (RCTs) on this subject have been published (Kakkar 2004; Klerk 2005). Therefore, we systematically reviewed How the intervention might work the literature to assess both efficacy and safety outcomes of par- enteral anticoagulation to prolong survival of patients with cancer.

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 6 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. OBJECTIVES • Major bleeding: we accepted the authors’ definitions of major bleeding. To evaluate the effectiveness and safety of parenteral anticoagu- • Minor bleeding: we accepted the authors’ definitions of lants in patients with cancer with no therapeutic or prophylactic minor bleeding. indication for anticoagulation. • Thrombocytopenia.

METHODS Search methods for identification of studies

Criteria for considering studies for this review Electronic searches Types of studies The search was part of a comprehensive search for studies of an- Randomized controlled trials. ticoagulation in patients with cancer. We conducted the original electronic search in January 2007 and updated it in February 2010. We searched the following databases: the Cochrane Central Regis- Types of participants ter of Controlled Trials (CENTRAL), Issue 1, 2010, MEDLINE Patients with cancer with no indication for prophylactic antico- (1966 onwards; accessed via OVID), EMBASE (1980 onwards; agulation (e.g. for acute illness, for central venous line placement, accessed via OVID) and ISI the Web of Science. The search strate- perioperatively) or for therapeutic anticoagulation (e.g. for the gies combined terms for anticoagulants, terms for cancer and a treatment of DVT or pulmonary embolism). search filter for RCTs. We used no language restrictions. We list the full search strategies for each of the electronic databases in Appendix 1. Types of interventions Experimental intervention: parenteral anticoagulants such as UFH, LMWH and fondaparinux. Searching other resources Comparator intervention: placebo or no intervention. We handsearched the conference proceedings of the American We also considered studies comparing different parenteral antico- Society of Clinical Oncology (ASCO, starting with its first volume, agulants. The protocol should have planned to provide all other 1982) and of the American Society of Hematology (ASH, starting co-interventions (e.g. chemotherapy) similarly. with its 2003 issue). We reviewed the reference lists of papers included this review and of other relevant systematic reviews. We used the related article feature in PubMed to identify additional Types of outcome measures articles. The outcome measures did not constitute criteria for including studies. Data collection and analysis Primary outcomes • All-cause mortality; pre-specified at 12 months, 24 months and over the duration of the trial. Selection of studies Two review authors independently screened the titles and abstracts of identified articles for eligibility. We retrieved the full text of Secondary outcomes articles judged as potentially eligible by at least one review au- • Symptomatic venous thromboembolism (DVT and/or thor. Two review authors then independently screened the full- pulmonary embolism): DVT events had to be diagnosed using text articles for eligibility using a standardized form with explicit an objective diagnostic test such as: venography, 125I- inclusion and exclusion criteria. The two review authors resolved fibrinogen-uptake test, impedance plethysmography or their disagreements by discussion or by consulting a third review compression ultrasound. Pulmonary embolism events had to be author. diagnosed using an objective diagnostic test such as: pulmonary perfusion/ventilation scans, computed tomography, pulmonary angiography or autopsy. Data extraction and management • Health-related quality of life: had to be measured using a We developed and used a standardized data extraction form. Two validated tool. authors independently extracted data from each included study

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 7 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. and resolved their disagreements by discussion. We aimed at col- • Type of cancer. lecting data related to the following. • Stage of cancer. • Time since cancer diagnosis. • Co-interventions including radiotherapy, chemotherapy Participants and hormonal therapy (type and duration). • Number of patients randomized to each treatment arm. • Use of indwelling central venous catheters. • Number of patients followed up in each treatment arm. • Number of withdrawals from treatment in each treatment arm. Interventions • Population characteristics (age, gender, co-morbidity). • Type of anticoagulant: UFH, LMWH or fondaparinux. • History of VTE (venous thromboembolism). • Dose: prophylactic versus therapeutic (Table 2).

Table 2. LMWH definitions of prophylactic and therapeutic dosages

LMWH Generic name Prophylactic dose Therapeutic dose

Lovenox Enoxaparin 40 mg once daily 1 mg/kg twice daily

Fragmin Dalteparin 2500 to 5000 units once daily 200 U/kg once daily or 100 U/kg twice daily

Innohep Tinzaparin, Logiparin 4500 units once daily 90 U/kg twice daily

Fraxiparine Nadroparin 35 to 75 anti-Xa international units/kg once 175 anti-Xa int. units/kg once daily daily

Certoparin Sandoparin 3000 anti-Xa international units once daily

• Duration of treatment. For dichotomous data, we extracted data necessary to conduct in- • Control: placebo or no intervention. tention-to-treat (ITT) analyses. We collected all-cause mortality at one year (time point defined a priori in the protocol) and at two years (time point defined post-hoc based upon results reported in Outcomes the individual RCTs). When we could not obtain the number of We extracted both time-to-event data (for the survival outcome) events at the time points of interest from the paper or from the au- and dichotomous data (for all outcomes). For mortality we col- thors, two review authors calculated these numbers independently lected data for the pre-specified time point of 12 months, but also and in duplicate from survival curves, if available (Altinbas 2004; for 24 months and for over the duration of follow up. Kakkar 2004). We used the mean of the two estimates when they For time-to-event survival data, we abstracted the log(hazard ratio differed. We assessed agreement between the two authors for each (HR)) and its variance from trial reports. If these were not reported, estimated value by calculating the percentage difference, which is we digitized the published Kaplan-Meier survival curves and es- the difference between the two estimates divided by the denomi- timated the log (HR) and its variance using Parmar’s methods nator (number of subjects at risk for the event) and multiplied by (Parmar 1998). We also noted the minimum and maximum dura- 100. tion of follow up, which are required to make these estimates. We We attempted to contact authors for incompletely reported data. performed these calculations in Stata 9, using a specially written We determined a priori to consider abstracts only if authors sup- program, which yielded the reported log(HR) and variance when plied us with full reports of their methods and results. used on the data presented in Table V of Parmar 1998 (Parmar 1998). Agreement between the two authors who extracted data from survival curves was high with an average percentage differ- Assessment of risk of bias in included studies ence of 2.5%.

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 8 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. First, we assessed risk of bias at the study level using the Cochrane compared the list of outcomes from those sources to the outcomes ’Risk of bias’ tool. Two review authors independently assessed the reported in the published paper. We also created an inverted funnel methodological quality of each included study and resolved their plot for the primary outcome of survival in order to check for disagreements by discussion. Methodological criteria included: possible publication bias. We did not create funnel plots for the • adequate sequence generation; other outcomes due to the low number of included trials for each • allocation concealment; outcome. • patient blinding; • provider blinding; • data collector blinding; Data synthesis • outcome assessor blinding; We calculated the agreement between the two independent review • analyst blinding; authors for the assessment of eligibility using a kappa statistic. • percentage of follow up and whether incomplete outcome For time-to-event data, we pooled the log(HRs) using a random- data were addressed; effects model (DerSimonian 1986) and the generic inverse vari- • whether the study was free of selective reporting; ance facility of RevMan 5 (RevMan 2008). • whether the study was stopped early for benefit; and For dichotomous data, we calculated the RR separately for each • whether the analysis followed the intention-to-treat (ITT) study. We then pooled the results of the different studies using a principle. random-effects model. Second, we assessed the quality of evidence at the outcome level us- ing the Grading of Recommendations Assessment, Development Subgroup analysis and investigation of heterogeneity and Evaluation (GRADE) approach (Cochrane Handbook). We planned to explore substantial heterogeneity by conducting subgroup analyses based on the type of intervention (LMWH Measures of treatment effect versus UFH) and the characteristics of participants (type, severity We collected and analyzed hazard ratios (HRs) for time-to-event and stage of cancer, and whether patients were on cancer treatment data and risk ratios (RRs) for dichotomous data. None of the or not). outcomes of interest was meta-analyzed as a continuous variable. Sensitivity analysis Unit of analysis issues We planned for sensitivity analyses excluding lower quality trials. The unit of analysis was the individual participant.

Dealing with missing data RESULTS All but two included studies reported 100% follow up. We ana- lyzed the available data assuming that any data that could be miss- Description of studies ing were missing at random. See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of ongoing studies. Assessment of heterogeneity We assessed heterogeneity between trials by visual inspection of forest plots, by estimation of the percentage heterogeneity between Results of the search trials which cannot be ascribed to sampling variation (Higgins The February 2010 search strategy identified a total of 8187 ci- 2003), and by a formal statistical test of the significance of the tations from which we removed the results of our January 2007 heterogeneity (Deeks 2001). If there was evidence of substantial search. In total, the title and abstract screening identified 67 po- heterogeneity, we investigated and reported the possible reasons tentially eligible citations. The full-text screening of the 65 cita- for this. tions identified nine eligible RCTs published either as full reports (Lebeau 1994; Altinbas 2004; Kakkar 2004; Klerk 2005; Sideras 2006; Weber 2008; Agnelli 2009; Perry 2010) or as an abstract Assessment of reporting biases with additional information provided by the author (Pelzer 2009). We assessed reporting bias by trying to identify whether the study We identified 10 published abstracts for six of the nine included was included in a trial registry, whether a protocol is available, RCTs (Altinbas 2004; Kakkar 2004; Klerk 2005; Sideras 2006; and whether the methods section provided a list of outcomes. We Pelzer 2009; Perry 2010). We also identified four eligible studies

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 9 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. published as five abstracts but for which we were not able to ob- duration of follow up were two and 33 months respectively. HRs tain the needed data from the authors: TOPIC 1 and 2 studies were estimated from published survival curves. (Freund 2003; Gatzemeier 2005; Haas 2005); Salat 1990; and Klerk et al. (MALTtrial) recruited 302 patients with different types Chazouilleres 1994. Agreement between authors for study eligi- of solid malignant tumors that could not be treated curatively and bility was excellent (kappa = 0.94). a minimum life expectancy of one month (Klerk 2005). Cancer types included: colorectal, breast, lung, gastric, oesophageal, liver, gallbladder, Katskin, prostate, pancreatic, cervical, urothelial, re- Included studies nal, ovarian, melanoma, endometrial and other cancers. Patients were randomized to receive either a LMWH (nadroparin; two The nine included studies recruited a total of 2857 participants weeks therapeutic dose followed by four weeks of a prophylactic (Agnelli 2009 included 1150 patients) and reported follow-up dose) or a placebo for six weeks without any concomitant chemo- data on 2843 (Lebeau 1994; Altinbas 2004; Kakkar 2004; Klerk therapy or radiotherapy. Study outcomes included mortality (at 2005; Sideras 2006; Agnelli 2009; Pelzer 2009). One study used six, months, one year and two years), major bleeding, and major or UFH as the intervention (Lebeau 1994) while the other eight used minor bleeding. Follow up was complete (100%). The minimum LMWH as the intervention (Altinbas 2004; Kakkar 2004; Klerk duration of follow up was not reported. The maximum duration 2005; Sideras 2006; Weber 2008; Agnelli 2009; Pelzer 2009; Perry of follow up was 84 months. The HR and its standard error were 2010). We did not identify any study using fondaparinux as the reported. intervention. Sideras et al. recruited 141 patients with different types of ad- Lebeau et al. recruited 277 patients with both limited and exten- vanced cancer and a minimum life expectancy of 12 weeks and sive small cell lung cancer (SCLC) 78% of which had a Karnof- ECOG state 0 to 2 (Sideras 2006). Patients were randomized ei- sky Performance Scale Index > 80 (Lebeau 1994). The Karnofsky ther to a prophylactic dose of a LMWH (dalteparin) or to placebo Performance Scale Index ranges from 0 (dead) to 100 (normal) or no intervention. Study outcomes included mortality (at one, (Karnofsky 1948). Patients were randomized to receive along with two and three years), VTE, and major bleeding. Follow-up data their chemotherapy either a prophylactic dose of UFH for five were available for 138 patients (98%). The minimum duration of weeks or no intervention. The study outcome was mortality (at follow up was not reported. The maximum duration of follow up one, two and three years). Follow up was complete (100%). The was 24 months. The authors supplied us with unpublished data minimum duration of follow up was not reported. The maximum giving the HR and its standard error. duration of follow up was 59 months. HRs were estimated from Perry et al. (PRODIGE trial) recruited 186 adults with newly diag- published survival curves, assuming all patients were followed up nosed malignant glioma (Perry 2010). Patients were randomized for 59 months. to receive a prophylactic dose of LMWH (dalteparin) or placebo. Kakkar et al. (the FAMOUS trial) recruited 385 patients with Study outcomes included objectively documented symptomatic advanced stage III or IV malignant disease of the breast, lung, DVT or pulmonary embolism (primary outcome), bleeding (ma- gastrointestinal tract, pancreas, liver, genitourinary tract, ovary or jor and all bleeding), quality of life, cognition assessments and uterus and a minimum life expectancy of three months (Kakkar death. The duration of follow up was 12 months. 2004). Patients were randomized to receive either a prophylactic Weber et al. recruited 20 patients aged 55 to 88 years with ad- dose of a LMWH (dalteparin) or placebo for 12 months with no vanced cancer and an estimated life expectancy of less than six restriction on concomitant chemotherapy or radiotherapy. The months (Weber 2008). Patients were randomized to receive ei- study outcomes included mortality (at one, two and three years), ther a prophylactic dose of LMWH (nadroparin) or no treatment. pulmonary embolism, DVT, major bleeding and minor bleeding. Study outcomes included death, venous thromboembolism and Follow-up data were available for 374 patients (97%). The min- major bleeding. imum duration of follow up was not reported. The maximum Pelzer et al. (CONKO 004 trial) recruited 312 chemotherapy- duration of follow up was 77 months. HRs were estimated from naive patients with histologically or cytologically confirmed ad- published survival curves, assuming all patients were followed up vanced pancreatic cancer (Pelzer 2009). Patients were randomized for 77 months. to receive or not to receive additional LMWH (enoxaparin) start- Altinbas et al. recruited 84 patients with both limited and exten- ing simultaneous to palliative systemic chemotherapy. Study out- sive SCLC and an Eastern Cooperative Oncology Group (ECOG) comes included symptomatic VTE, overall survival, major bleed- state < 3 (Altinbas 2004). The ECOG Performance Status scale ing and time to progression (TTP). The median duration of follow ranges from 0 (fully active) to 5 (dead) (Oken 1982). Patients up was 30.4 weeks. The results for VTE and major bleeding out- were randomized to receive either a prophylactic dose of a LMWH comes have been reported whereas the results for overall survival (dalteparin) or placebo for 18 weeks or less in combination with and TTP are still pending (will be presented at DGHO meeting chemotherapy in case of disease progression. Study outcomes in- in October 2010 as per the authors). cluded mortality (at one and two years), DVT and minor bleeding. Agnelli et al. (PROTECHT trial) recruited 1150 ambulatory Follow up was complete (100%). The minimum and maximum

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 10 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. patients with metastatic or locally advanced lung, gastrointesti- et al. reported a 94% follow-up rate for thromboembolic events nal, pancreatic, breast, ovarian or head and neck cancer (Agnelli and 87% follow-up rate for survival (Pelzer 2009). The remaining 2009). Patients were randomized to receive a prophylactic dose studies reported a 100% follow-up rate of nadroparin (3800 IU anti-Xa once a day) or placebo. The study treatment was started on the same day as chemotherapy (the first cycle or a new course) for a maximum of four months. Selective reporting The primary efficacy outcome was the composite of symptomatic The outcomes listed in the methods section were reported in the DVT of lower or upper limbs, pulmonary embolism (PE), vis- results section for seven studies (Lebeau 1994; Kakkar 2004; Klerk ceral or cerebral venous thrombosis, acute myocardial infarc- 2005; Sideras 2006; Weber 2008; Agnelli 2009; Pelzer 2009). One tion, ischemic stroke, acute peripheral arterial thromboembolism, study reported on all outcomes except for two listed in the meth- and unexplained death of possible VTE origin occurring during ods section (quality of life and cognition assessment) (Perry 2010). the study treatment plus 10 days. The secondary efficacy out- The outcomes were not listed in the methods section for one study comes were asymptomatic thromboembolic events incidentally di- (Altinbas 2004). Three studies were registered with ClinicalTri- agnosed, survival at the end of study treatment and at 12 months, als.gov (Agnelli 2009; Pelzer 2009; Perry 2010). Only one study superficial thrombophlebitis of the lower limbs, response to che- had a published protocol and did report on all outcomes described motherapy and, for patients with central-venous-catheter (CVC), in that protocol (Pelzer 2009). CVC-related complications of possible thrombotic origin.

Other potential sources of bias Excluded studies Only one study was stopped early by patient monitoring commit- We excluded 43 articles from this review for the following reasons: tee for insufficient accrual (Sideras 2006). intervention was topical heparin (n = 1) or intraportal infusion We judged that in the study by Lebeau et al. (Lebeau 1994) pa- of heparin (n = 2); intervention consisted of only two doses of tients received similar co-interventions although brain and tho- LMWH (n = 1); comparison was of heparin versus vitamin K racic irradiation depended on response to treatment. In that study antagonist (n = 3); studies included no cancer patients (n = 2); no 11% and 7% respectively of patients randomized to heparin and survival outcome (n = 1); study design was not a RCT (n = 16); control groups received radiotherapy. letter to the editor or editorial (n = 12); publication was a review All but one study reported using the intention-to-treat (ITT) anal- (n = 5). ysis principle (Sideras 2006).

Risk of bias in included studies Effects of interventions See: Summary of findings for the main comparison heparin compared to no heparin for patients with cancer who have no Allocation other therapeutic or prophylactic indication for anticoagulation Allocation was adequately concealed in eight of the nine studies (Lebeau 1994; Kakkar 2004; Klerk 2005; Sideras 2006; Weber All-cause mortality 2008; Agnelli 2009; Pelzer 2009; Perry 2010) and it was not con- cealed in one study (Altinbas 2004). Mortality at 12 months Meta-analysis of the eight RCTs, including 2531 participants, Blinding found no statistically significant difference in the risk of death at While four studies blinded patients, providers, data collectors 12 months between heparin and placebo (RR 0.93; 95% CI 0.85 and outcome adjudicators (Kakkar 2004; Klerk 2005; Agnelli to 1.02) (see Analysis 1.1). The percentage of the variability in 2009; Perry 2010), one study blinded only data analysts (Pelzer effect estimates that is due to heterogeneity rather than sampling 2009), and four studies did not follow any blinding (Lebeau 1994; error (chance) may represent moderate heterogeneity (I2 = 35%). Altinbas 2004; Sideras 2006; Weber 2008) The associated risk of bias is presented in Figure 1. The qual- ity of evidence was moderate (Summary of findings for the main comparison). The inverted funnel plot for the primary outcome Incomplete outcome data of mortality at one year did not suggest publication bias, but there Agenelli et al. reported a 98.5% follow-up rate (Agnelli 2009). were relatively few trials to permit an accurate assessment (Figure Sideras et al. reported a 98% follow-up rate (Sideras 2006). Pelzer 2).

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 11 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Figure 1. ’Risk of bias’ summary - mortality at 12 months: review authors’ judgements about each methodological quality item for each included study that evaluated mortality

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 12 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Figure 2. Funnel plot of comparison: 1 Heparin vs placebo, outcome: 1.1 Mortality at 12 months.

In a subgroup analysis of patients with small cell lung cancer (SCLC) (Altinbas 2004; Lebeau 1994) versus other types of can- cer, the test for subgroup effect was statistically significant (P = 0.03) (Figure 3): RR 0.86 (95% CI 0.75 to 0.98) for SCLC versus RR 0.96 (95% CI 0.86 to 1.07) for other types of cancer.

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 13 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Figure 3. Subgroup analysis comparing SCLC with other types of cancer for mortality at 12 months

In a subgroup analysis of patients with advanced cancer (includ- ing patients with extensive SCLC) versus patients with non-ad- Seven studies, including 1381 participants, reported data allowing vanced cancer (including patients with limited SCLC), we found their inclusion in the time to event meta-analysis. Meta-analysis no significant difference between the effects of heparin in the two indicated that heparin was associated with a statistically significant subgroups (P = 0.51). reduction in the risk of death compared to placebo (RR 0.79; 95% CI 0.67 to 0.93) (see Analysis 1.3). The percentage of the vari- Mortality at 24 months ability in effect estimates that is due to heterogeneity rather than sampling error (chance) may represent moderate heterogeneity (I In a post-hoc meta-analysis of five RCTs, including 1175 partici- 2 = 44%). pants, heparin was associated with a statistically significant reduc- tion in the risk of death at 24 months compared to placebo (RR 0.92; 95% CI 0.88 to 0.97) (see Analysis 1.2). The percentage Venous thromboembolism of the variability in effect estimates that is due to heterogeneity Meta-analysis of seven RCTs, including 2264 participants, found 2 rather than chance is not important (I = 0%). that heparin was associated with a statistically significant reduction The test for subgroup effect was not statistically significant for ei- in the risk of symptomatic VTE compared to placebo (RR 0.55; ther of the two subgroup analyses (advanced versus non-advanced 95% CI 0.37 to 0.82) (see Analysis 1.4). The percentage of the and SCLC versus non SCLC). variability in effect estimates that is due to heterogeneity rather than chance is not important (I2 = 0%). The associated risk of bias is presented in Figure 4. The quality of evidence was high Mortality - time-to-event analysis (Summary of findings for the main comparison).

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 14 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Figure 4. ’Risk of bias’ summary - venous thromboembolism: review authors’ judgements about each methodological quality item for each included study that evaluated mortality

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 15 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Major bleeding Meta-analysis of the nine RCTs, including 2843 participants, found that heparin was associated with a statistically significant increase in the risk of major bleeding compared to placebo (RR 1.30; 95% CI 0.59 to 2.88) (see Analysis 1.5). The percentage of the variability in effect estimates that is due to heterogeneity rather than sampling error (chance) may represent moderate heterogene- ity (I2 = 24%). The associated risk of bias is presented in Figure 5. The quality of evidence was moderate (Summary of findings for the main comparison).

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 16 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Figure 5. ’Risk of bias’ summary - major bleeding: review authors’ judgements about each methodological quality item for each included study that evaluated mortality

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 17 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Minor bleeding Meta-analysis of seven RCTs, including 2345 participants, found no statistically significant difference in the risk of minor bleeding between heparin and placebo (RR 1.05; 95% CI 0.75 to 1.46) (see Analysis 1.6). The percentage of the variability in effect estimates that is due to heterogeneity rather than chance is not important (I 2 = 0%). The associated risk of bias is presented in Figure 6. The quality of evidence was moderate (Summary of findings for the main comparison).

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 18 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Figure 6. ’Risk of bias’ summary - minor bleeding: review authors’ judgements about each methodological quality item for each included study that evaluated mortality

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 19 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Quality of life One study (Sideras 2006) assessed quality of life using the Uniscale and the Symptom Distress Scale (SDS). The authors reported that the scores for the two scales were similar for the two study groups, both at baseline and at follow up. The associated risk of bias is presented in Figure 7. The quality of evidence was low (Summary of findings for the main comparison).

Figure 7. ’Risk of bias’ summary - quality of life: review authors’ judgements about each methodological quality item for each included study that evaluated mortality

DISCUSSION Thrombocytopenia

Summary of main results Four studies, including 683 participants, assessed thrombocytope- The effect of parenteral anticoagulation (with either unfraction- nia as an outcome and reported the occurrence of no events ated heparin (UFH) or low molecular weight heparin (LMWH)) (Lebeau 1994; Altinbas 2004; Klerk 2005; Weber 2008). on mortality of patients with cancer who have no therapeutic or Sensitivity analyses prophylactic indication for anticoagulation was statistically and The sensitivity analysis excluding the lower quality study (Altinbas clinically significant at 24 months but not at 12 months. While 2004) from the analyses did not change the results significantly. parenteral anticoagulation was associated with a reduction in ve-

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 20 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. nous thromboembolic events, it was not associated with a statisti- subgroup analyses to explore effect modifiers such as type and stage cally significant effect on major bleeding, minor bleeding or qual- of cancer, and dosing, schedules and duration of heparin therapy. ity of life. We did not identify any study using fondaparinux as The interpretation of findings is also limited by not including data the anticoagulant. from the trials published as abstracts only. Also, for two studies The discrepancy between the effect on mortality at 12 and 24 (Altinbas 2004; Kakkar 2004), we had to calculate the number of months might be related to the fact that the two analyses included mortal events at 12 and 24 month from the survival curves. different studies with populations varying by the type and stage of cancer. In fact, the subgroup analysis for mortality at 12 months was statistically significant and suggested survival benefit in pa- Agreements and disagreements with other tients with small cell lung cancer (SCLC) but not in patients with studies or reviews other types of cancer. The second subgroup analysis suggested but did not confirm that patients with advanced cancer might not ben- Lazo-Langner et al. conducted a systematic review addressing the efit from parenteral anticoagulation. These suggestions are con- same question as this review (Lazo-Langner 2007). Although that sistent with the findings of Klerk et al (Klerk 2005): the hazard review had different inclusion criteria from our review (in particu- ratio for survival was 0.64 (95% CI 0.45 to 0.90) for patients with lar, it excluded the trial of Lebeau) and obtained slightly different life expectancy greater than six months and 0.88 (95% CI 0.62 to estimates of HRs using Parmar’s methods, it reported a HR com- 1.25) for patients with life expectancy shorter than six months. paring mortality in the heparin and control arms of 0.83 (95% CI 0.70 to 0.99), consistent with our findings. The authors did not report any subgroup analysis. Overall completeness and applicability of Kuderer conducted a systematic review on the efficacy and safety evidence of parenteral anticoagulation as cancer treatment (Kuderer 2007). The authors analyzed the study by Lebeau et al. separately and The included studies recruited patients with variety of cancer types found a RR for one-year mortality of 0.88 (95% CI 0.79 to 0.98); and stages which should increase the applicability of the results. the RR for major bleeding was 1.68 (95% CI 0.86 to 3.27). In a The results apply best to LMWH given only one study evaluated meta-regression analysis, the authors found that only tumor type UFH. Unfortunately, not enough data were available to evaluate significantly influenced mortality risk reduction with significant the impact of the intervention on bleeding outcomes or on quality reduction in patients with SCLC. of life. The latter outcome is important given the potential burden Our current findings differ from those of the two above systematic of daily subcutaneous injections. reviews as well as from our previous findings (prior to this update) As mentioned above, we identified four eligible studies published in that mortality at one year is not statistically significant. This as five abstracts but for which we were not able to obtain the finding was mainly driven by the results of the largest and most needed data from the authors. The TOPIC 1 and 2 studies re- recently published trial (Agnelli 2009), the inclusion of which spectively included 353 patients with advanced cancer and 547 introduced heterogeneity into the results of mortality at 12 months patients with non-small cell lung carcinoma and reported no sur- (I2 increased from 0% to 40%). This trial included patients with vival benefit with LMWH (Freund 2003; Gatzemeier 2005; Haas metastatic or locally advanced cancer of different types. As noted 2005). Chazouilleres 1994 recruited 51 patients with unresectable previously, the subgroup analysis conducted by Klerk 2005 as well hepato cellular carcinoma and reported a lower short-term mor- as of our subgroup analyses, although not statistically significant, tality rate with LMWH. Salat 1990 did not report on mortality suggest that the type and stage of cancer may modify the effect outcome. parenteral anticoagulation. The results of this review, based on four additional RCTs (Weber 2008; Agnelli 2009; Pelzer 2009; Perry 2010) compared to the Quality of the evidence prior review that included five RCTs, indicate that the effect on Our systematic approach to searching, study selection and data ex- mortality may not be present, smaller or limited to specific sub- traction should have minimized the likelihood of missing relevant groups. studies. The quality of evidence was high for symptomatic venous thromboembolism, moderate for mortality, major bleeding and minor bleeding, and low for quality of life. AUTHORS’ CONCLUSIONS

Potential biases in the review process Implications for practice The small number of studies and the inclusion of different types This systematic review found a potential survival benefit from hep- of cancer in the same study precluded us from conducting the arin therapy in cancer patients. We also found a significant de-

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 21 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. crease in thrombotic events with potentially no increase in bleed- advanced) of cancers using different types, dosing, schedules and ing events. duration of therapy (Alifano 2004). Some of the ongoing studies may provide such information (ABEL; FOCUS; FRAGMATIC; The decision for a patient with cancer to start heparin therapy in GASTRANOX; INPACT; NCT00916669). Also, an individual the absence of a therapeutic or a prophylactic indication should data meta-analysis may be particularly useful in clarifying how the balance the benefits and downsides and integrate the patient’s val- type and stage of cancer modify the effect of parenteral anticoag- ues and preferences (Haynes 2002). Patients with a high prefer- ulation. ence for a potential survival prolongation, limited aversion to po- tential bleeding, and who do not consider heparin (both UFH or LMWH) therapy a burden may opt to use heparin, while those with aversion to bleeding may not. ACKNOWLEDGEMENTS We thank Ms. Ann Grifasi for her administrative support. We Implications for research thank Dr. Loprinzi and Dr. Paul Novotny of the Mayo Clinic for There is a need to understand the effects of heparin (including supplying additional data relating to the study of Sideras 2006. We UFH and LMWH) and other anticoagulants in patients with dif- also thank Dr. Lebeau, Dr. Altinbas and Dr. Pelzer for supplying ferent types (SCLC versus others) and stages (advanced versus not additional data.

REFERENCES

References to studies included in this review Klerk 2005 {published data only} Klerk CP, Smorenburg SM, Otten HM, Lensing AW, Prins MH, Piovella F, et al.The effect of low molecular weight Agnelli 2009 {published data only} heparin on survival in patients with advanced malignancy. Agnelli G, Gussoni G, Bianchini C, Verso M, Mandalà M, Journal of Clinical Oncology 2005;23(10):2130–5. Cavanna L, et al.PROTECHT Investigators. Nadroparin Klerk CPW, Smorenburg SM, Otten JMMB, Buller HR. for the prevention of thromboembolic events in ambulatory Malignancy and low-molecular weight-heparin therapy: the patients with metastatic or locally advanced solid cancer MALT trial. Journal of Thrombosis & Haemostasis, Abstracts receiving chemotherapy: a randomised, placebo-controlled, from XIX International ISTH Congress July 2003;1(Suppl 1): double-blind study. Lancet Oncology 2009;10(10):943–9. Abstract number: OC195. Altinbas 2004 {published data only} Klerk CPW.Smorenburg SM, Otten JMMB, Buller HR. Altinbas M, Coskun HS, Ozkan M, Eser B, Unal A, Malignancy and low molecular weight-heparin therapy: the Cetin M, et al.A randomized clinical trial of combination MALT trial. Pathophysiology of Haemostasis and Thrombosis chemotherapy with and without low molecular weight 2003;33(Suppl 1):75. heparin in small cell lung cancer. Journal of Thrombosis and Lebeau 1994 {published data only} Haemostasis 2004;2:1266–71. Lebeau B, Chastang C, Brechot JM, Capron F, Dautzenberg Altynbas M, Coskun HS, Er O, Ozkan M, Eser B, Delaisements C, et al.Subcutaneous heparin treatment B, Unal A, et al.Prospective randomized study of increases survival in small cell lung cancer. “Petites Cellules” epirubicine cyclophosphamide and vincristine combination Group. Cancer 1994;74(1):38–45. chemotherapy (CEV): low molecular weight heparin (LMWH) in small cell lung cancer (SCLC). Proceedings of Pelzer 2009 {published data only} the American Society of Clinical Oncology. 2001; Vol. 20: Pelzer U, Deutschinoff G, Opitz B, Stauch M, Reitzig 1280. P, Hahnfeld S, et al.A prospective, randomized trial of Altynbas M, Ozkan M, Coskun HS, Er O, Eser B, Unal A, simultaneous pancreatic cancer treatment with enoxaparin et al.Efficiency of cyclophosphamide, epirubicin, vincristine and chemotherapy - first results of the CONKO 004 trial. (CEV) +/- low molecular weight heparin (LMWH) in small Onkologie - DGHO meeting Oct 2009;580:abstract. cell lung cancer (SCLC): preliminary results. Annals of Pelzer U, Hilbig A, Stieler J, Roll L, Riess H, Dorken Oncology. 2000; Vol. 11:117. B, et al.A prospective, randomized trial of simultaneous pancreatic cancer treatment with enoxaparin and Kakkar 2004 {published data only} chemotherapy (PROSPECT - CONKO 004). Onkologie Kakkar AK, Levine MN, Kadziola Z, Lemoine NR, Low V, 2005;28(Suppl 3):54 (abstract 151). Patel HK, et al.Low molecular weight heparin, therapy with Pelzer U, Oettle H, Stauch M, Opitz B, Stieler J, Scholten dalteparin, and survival in advanced cancer: the fragmin T, et al.Prospective, randomized open trial of enoxaparin in advanced malignancy outcome study (FAMOUS). Journal patients with advanced pancreatic cancer undergoing first- of Clinical Oncology 2004;22(10):1944–8. line chemotherapy. XXIst Congress of the International

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 22 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Society on Thrombosis and Haemostasis. July 6–12, 2007; Barkagan 1997 {published data only} Vol. Geneva:Abstract no P-T-488. Barkagan ZS. The results of the use of low molecular Riess H, Pelzer U, Hilbig A, Stieler J, Opitz B, Scholten T, weight heparin (LMWH) for prevention and treatment et al.Rationale and design of PROSPECT-CONKO 004: of thrombosis in cancer patients. Thrombosis and a prospective, randomized trial of simultaneous pancreatic Haemostasis. 1997:772. cancer treatment with enoxaparin and chemotherapy). Bitsch 1990 {published data only} BMC Cancer 2008;8:361. Bitsch M, Hermann GG, Andersen JP, Steven K. Low Perry 2010 {published data only} dose intravesical heparin as prophylaxis against recurrent Perry JR, Julian JA, Laperriere NJ, Geerts W, Agnelli noninvasive (stage Ta) bladder cancer. Journal of Urology G, Rogers LR, et al.PRODIGE: a randomized placebo- 1990;144(3):635–6. controlled trial of dalteparin low molecular weight heparin Blaszczyk 1970 {published data only} (LMWH) thromboprophylaxis in patients with newly Blaszczyk M, Ursyn-Niemcewicz W, Pawlak F. Heparin diagnosed malignant glioma. Journal of Thrombosis and precipitable fraction (HPF) in malignant tumors of the Hemostasis 2010;8(9):1959-65. respiratory tract. Gruzlica i Choroby Pluc 1970;38(4): Perry JR, Rogers L, Laperriere N, Julian J, Geerts W, 321–8. Agnelli G et al for the Ontario Clinical Oncology Group Buckman 2005 {published data only} and PRODIGE Investigators. PRODIGE: a phase III Buckman RA, Wong NS, Clemons M, Verma S, Trudeau randomized placebo-controlled trial of thromboprophylaxis ME, Roche K, et al.Phase I-II study of DaICM-P [daily using dalteparin low molecular weight heparin (LMWH) in dalteparin (Dal), cyclophosphamide (C) and prednisone (P) patients with newly diagnosed malignant glioma. Journal of and bi-weekly methotrexate (M)] as therapy for metastatic Clinical Oncology. 2007; Vol. 25:77s (abstract 2011). breast cancer (MBC). Journal Of Clinical Oncology 2005;23 Sideras 2006 {published data only} (16):52S. Sideras K, Schaefer PL, Okuno SH. Low-molecular-weight Craven 2001 {published data only} heparin in patients with advanced cancer: a phase 3 clinical Craven R. Heparin and cancer revisited. Trends in trial. Journal of Vascular Surgery 2007;45(4):861. Pharmacological Sciences 2001;22(5):1. Sideras K, Schaefer PL, Okuno SH, Sloan JA, Kutteh L, Dakhil SR, et al.Phase III clinical trial evaluating low- Crossno 2009 {published data only} molecular weight heparin (LMWH) in patients with Crossno RJ. Anticoagulation for the long-term treatment of advanced cancer: A North Central Cancer Treatment Group venous thromboembolism in patients with cancer. Journal study. Journal of Clinical Oncology 2005;23(16):775S. of Pain and Palliative Care Pharmacotherapy 2009;23(1): Sideras K, Schaefer PL, Okuno SH, Sloan JA, Kutteh L, 65–6. Fitch TR, et al.Low-molecular-weight-heparin in patients Di Nisio 2005 {published data only} with advanced cancer: a phase 3 clinical trial. Mayo Clinic Di Nisio M, Buller HR, Porreca E. Do low-molecular- Proceedings 2006;81(6):758–67. weight heparins improve the survival of cancer patients?. Weber 2008 {published data only} Nature Clinical Practice Oncology 2005;2(12):612–3. Weber C, Merminod T, Herrmann FR, Zulian GB. Edlis 1976 {published data only} Prophylactic anti-coagulation in cancer palliative care: a Edlis HE, Goudsmit A, Brindley C, Niemetz J. Trial prospective randomised study. Support Care Cancer 2008; of heparin and cyclophosphamide (NSC-26271) in the 16(7):847–52. treatment of lung cancer. Cancer Treatment Report 1976;60 (5):575–8. References to studies excluded from this review Elias 1972 {published data only} Elias EG, Brugarol A. Role of heparin in chemotherapy of Alifano 2005 {published data only} solid tumors - preliminary clinical trial in carcinoma of lung. Alifano M, Maggiore O, Benedetti G, Trisolini R. Can low- Cancer Chemotherapy Reports Part 1 1972;56(6):783–5. molecular-weight heparin improve the outcome of patients with operable non-small cell lung cancer? An urgent call for Elias 1973a {published data only} research. Chest 2004;126:601–607. Elias EG. Heparin as an adjuvant to chemotherapy In lung Arbit 2005 {published data only} carcinoma. Proceedings of the American Association for Arbit E, Alifano M, Maggiore O, Benedetti G, Trisolini R. Cancer Research. 1973; Vol. 14:26. Low-molecular-weight heparin and outcomes. Chest 2005; Elias 1973b {published data only} 128(1):471–2. Elias EG, Sepulveda F, Mink IB. Increasing the efficiency of cancer chemotherapy with heparin: “clinical study”. Journal Barberi-Heyob 1995 {published data only} of Surgical Oncology 1973;5(2):189–93. Barberi-Heyob M, Merlin JL, Vigneron M, Conroy T. Addition of heparin in 5-fluorouracil solution for portal vein Elias 1973c {published data only} infusion has no influence on its stability under clinically Elias EG. Heparin therapy combined with chemotherapy in relevant conditions. Anti-Cancer Drugs 1995;6(1):163–4. metastatic cancer. Cancer Bulletin 1973;25(6):116–9.

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 23 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Elias 1974 {published data only} Loynes 2002 {published data only} Elias EG. Heparin anticoagulation as adjuvant to Loynes JT, Zacharski LR, Rigas JR. Regression of metastatic chemotherapy in carcinoma of the lung. Journal of Medicine non-small cell lung cancer with low molecular weight 1974;5(1):114–32. heparin. Thrombosis and Haemostasis 2002;88(4):686. Elias 1975 {published data only} Lykke 2003 {published data only} Elias EG, Shukla SK, Mink IB. Heparin and chemotherapy Lykke J, Rasmussen HM, Nielsen HJ. Heparin as adjuvant in the management of inoperable lung carcinoma. Cancer in the treatment of colorectal cancer?. Ugeskrift for Laeger 1975;36(1):129–36. 2003;165(18):1866–7. Fielding 1992 {published data only} Mammen 2004 {published data only} Fielding LP, Hittinger R, Grace RH, Fry JS. Randomized Mammen EF. Expanded role of low-molecular-weight controlled trial of adjuvant chemotherapy by portal- heparins in hematologic and oncologic indications. vein perfusion after curative resection for colorectal Seminars in Thrombosis and Hemostasis 2004;30(Suppl 1): adenocarcinoma. Lancet 1992;340(8818):502–6. 1–2. Graf 1994 {published data only} Meyer 2007 {published data only} Graf B, Graf AH, Traun H, Forstner K, Rettenbacher Meyer G. Does low-molecular-weight heparin influence L, Sailer S, et al.Prophylaxis of thromboembolism in cancer-related mortality. Annals of Oncology 2007;18(3): radiotherapy for gynecologic malignancies: low molecular 609–10. weight (LMW) heparin (fragmin (R)) vs coumarin (sintrom Mousa 2001 {published data only} (R)). Haemostasis 1994;24(Suppl 1):315 - Abstract No 6. Mousa SA, Mohammed S. Anti-angiogenesis mechanisms Graf 1996 {published data only} and efficacy of the low molecular weight heparin, tinzaparin: Graf AH, Graf B, Traun H, Staudach A. Risk and prevention anti-cancer efficacy beyond its anticoagulants. Blood 2001; of thromboembolism complications in gynecologic 98(11):181B. malignancies. Gynakologisch-Geburtshilfliche Rundschau Munstedt 1996 {published data only} 1996;36(1):37–9. Munstedt K, Kemkes-Matthes B, Matthes KJ, Vahrson Green 1992 {published data only} H. The behavior of the activation parameters of plasma Green D, Hull RD, Brant R, Pineo GF. Lower mortality in coagulation under HDR-afterloading therapy in patients cancer patients treated with low-molecular-weight versus with endometrial carcinoma (German). Strahlentherapie standard heparin. Lancet 1992;339(8807):1476. und Onkologie 1996;172(1):39–42. Guimbretiere 1982 {published data only} Nash 2000 {published data only} Guimbretiere J, Raffi F, Dabouis G. Heparin therapy in Nash G. Heparin for colorectal cancer. Journal of the Royal hypercoagulable state of lung cancer patients. Haemostasis Society of Medicine 2000;93(10):554. 1982;12(1-2):139. Nishioka 2007 {published data only} Jorgensen 2001 {published data only} Nishioka J, Goodin S. Low-molecular-weight heparin Jorgensen LN. Anti-Xa levels in plasma predict late survival in cancer-associated thrombosis: treatment, secondary in patients after major surgery for malignancy receiving prevention, and survival. Journal of Oncology Pharmacy prophylaxis with low molecular weight heparins. Blood Practice 2007;13(2):85–97. 2001;98:43a. Nitti 1997 {published data only} Kohanna 1983 {published data only} Nitti D, Wils J, Sahmoud T, Curran D, Couvreur ML, Lise Kohanna FH, Sweeney J, Hussey S. Effect of perioperative M, et al.Final results of a phase III clinical trial on adjuvant low-dose heparin administration on the course of colon intraportal infusion with heparin and 5-fluorouracil (5-FU) cancer. Surgery 1983;93(3):433–8. in resectable colon cancer (EORTC GITCCG 1983-1987). Lecumberri 2005 {published data only} European Organization for Research and Treatment of Cancer Lecumberri R, Paramo JA, Rocha E. Anticoagulant 1997;33(8):1209–15. treatment and survival in cancer patients. The evidence Retik 1962 {published data only} from clinical studies. Haematologica 2005;90(9):1258–66. Retik AB, Arons MS, Ketcham AS, Mantel N. The effect Lemoine 2005 {published data only} of heparin on primary tumors and metastases. Journal of Lemoine NR. Antithrombotic therapy in cancer. Journal of Surgical Research 1962;2:49–53. Clinical Oncology 2005;23(10):2119–20. Rohwedder 1977 {published data only} Levine 2005 {published data only} Rohwedder JJ, Sagastume E. Heparin and polychemotherapy Levine MN, Lee AYY, Kakkar AK. Low-molecular-weight for treatment of lung cancer. Cancer Treatment Report 1977; heparin versus oral anticoagulant therapy for the long-term 61(7):1399–1401. treatment of symptomatic venous thromboembolism: is Siragusa 1999 {published data only} there any difference in cancer-related mortality? Reply. Siragusa S. Low molecular weight heparins could be Journal of Clinical Oncology 2005;23(28):7250. important in cancer. BMJ 1999;319(7213):851.

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 24 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Spigel 2005 {published data only} Additional references Spigel DR. Low-molecular-weight heparin improves survival in patients with cancer. Journal of Clinical Outcomes Alifano 2004 Management 2005;12(5):241–2. Alifano M, Benedetti G, Trisolini R. Can low-molecular- Von Hugo 1981 {published data only} weight heparin improve the outcome of patients with Hugo R, Hafter R, Hiller KF,Lochmuller H, Selbmann HK, operable non-small cell lung cancer?: An urgent call for Graeff H. Prevention of deep vein thrombosis in patients research. Chest 2004;126:601–7. with gynaecologic cancer undergoing radiotherapy. A Chazouilleres 1994 comparison of calcium-heparin and semi-synthetic heparin Chazouilleres O, Poupon R, Gatineausaillant G, Roulot analogue [Thromboembolische Komplikationen waehrend D, Barbare JC, Labadie H, et al.Beneficial effect of low der Strahlenbehandlung gynaekologischer karzinome]. molecular weight heparin (fraxiparin) on short term Geburtshilfe und Frauenheilkunde 1981;41(3):179–83. mortality in patients with unresectable hepatocellular Wojtukiewicz 2003 {published data only} carcinoma (HCC). A randomized study. Gastroenterology Wojtukiewicz MZ, Kozlowski L, Ostrowska K, Dmitruk A, 1994;106(4):A874. Zacharski LR. Low molecular weight heparin treatment for Cochrane Handbook malignant melanoma: a pilot clinical trial. Thrombosis and Higgins JPT, Green S (editors). Cochrane Handbook for Haemostasis 2003;89(2):405–7. Systematic Reviews of Interventions Version 5.0.2 [updated Zacharski 2003 {published data only} September 2009]. The Cochrane Collaboration, 2008. Zacharski LR. Heparin treatment of malignancy: the case Available from www.cochrane-handbook.org. for clinical trials in colon cancer. Thrombosis Research 2003; 110(4):213–4. Deeks 2001 Deeks JJ, Altman DG, Bradburn MJ. Statistical methods References to ongoing studies for examining heterogeneity and combining results from several studies in meta-analysis. In: Egger M, Davey Smith ABEL {published data only} G, Altman DG editor(s). Systematic Reviews in Health Care: Adjuvant BEmiparin in Small Cell Lung Carcinoma (ABEL Meta-Analysis in Context. 2nd Edition. London: BMJ STUDY). Ongoing study June 2005. Publication Group, 2001. FOCUS {published data only} DerSimonian 1986 A Phase II Randomized Study of Fragmin in Ovarian DerSimonian R, Laird N. Meta-analysis in clinical trials. Cancer: Utility on Survival (FOCUS). Ongoing study Controlled Clinical Trials 1986;7:177–88. October 2005. FRAGMATIC {published data only} Dvorak 1986 Griffiths GO, Burns S, Noble SI, Macbeth FR, Cohen Dvorak HF. Tumors: wounds that do not heal. Similarities D, Maughan TS. FRAGMATIC: a randomised phase III between tumor stroma generation and wound healing. New clinical trial investigating the effect of fragmin added to England Journal of Medicine 1986;315(26):1650–9. standard therapy in patients with lung cancer. BMC Cancer Francis 1998 2009;9:355. Francis JL, Biggerstaff J, Amirkhosravi A. Hemostasis and GASTRANOX {published data only} malignancy. Seminars in Thrombosis and Hemostasis 1998; Overall Survival of Inoperable Gastric/GastroOesophageal 24(2):93–109. Cancer Subjects on Treating With LMWH + Chemotherapy (CT) vs Standard CT (GASTRANOX). Ongoing study July Freund 2003 2008. Freund M, Kakkar AK, Haas S, Heilmann L, von Tempelhoff GF, Brom J, et al.A randomized trial of the INPACT {published data only} low molecular weight heparin certoparin against placebo in The INPACT Study (Improving With Nadroparin the the long-term prevention of venous thromboembolism in Prognosis in Advanced Cancer Treatment). A Randomized, patients with metastatic breast cancer. Blood 2003;102(11): Controlled Trial to Evaluate the Effects of Nadroparin on 210A. Survival and Disease Progression in Patients With Advanced Malignancies of the Lung, Pancreas, or Prostate. Ongoing Gatzemeier 2005 study May 2006. Gatzemeier U, Freund M, Haas S, Kakkar A, Zatloukai P, NCT00916669 {published data only} Kelbel C, et al.Prevention of thromboembolic complications A Randomized Phase II Study to Evaluate the Effect of Two with the low-molecular-weight heparin certoparin in non- Different Doses of Enoxaparin Sodium in Combination small-cell lung carcinoma (TOPIC-2). Lung Cancer. 2005; With Standard Chemotherapy (Cisplatin Plus Etoposide) Vol. 49:S56. With Respect to Time to Tumor Progression (TTP) in Girolami 2006 Patients With Newly Diagnosed Extensive Stage Small Girolami B, Girolami A. Heparin-induced Cell Lung Cancer (SCLC) Without Underlying Venous thrombocytopenia: a review. Seminars in Thrombosis and Thromboembolism. Ongoing study July 2008. Hemostasis 2006;32(8):803–9.

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 25 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Haas 2005 Oken 1982 Haas SK, Kakkar AK, Kemkes-Matthes B, Freund M, Oken MM, Creech RH, Tormey DC, Horton J, Davis TE, Gatzemeier U, Heilmann L, et al.Prevention of venous McFadden ET, et al.Toxicity and response criteria of the thromboembolism with low-molecular-weight heparin in Eastern Cooperative Oncology Group. American Journal of patients with metastatic breast or lung cancer - results of the Clinical Oncology 1982;5(6):649–55. TOPIC studies. Journal of Thrombosis & Haemostasis 2005; Parmar 1998 3(1):Abstract number: OR059. Parmar MKB, Torri V, Stewart L. Extracting summary Haynes 2002 statistics to perform meta-analyses of the published Haynes RB, Devereaux PJ, Guyatt GH. Clinical expertise literature for survival endpoints. Statistics in Medicine 1998; in the era of evidence-based medicine and patient choice. 17:2815–34. Vox Sanguinis 2002;83(Suppl 1):383–6. Prandoni 1992 Prandoni P, Lensing AW, Buller HR, Carta M, Cogo A. Higgins 2003 Comparison of subcutaneous low-molecular-weight heparin Higgins JP, Thompson SG, Deeks JJ, Altman DG. with intravenous standard heparin in proximal deep-vein Measuring inconsistency in meta-analyses. BMJ 2003;327 thrombosis. Lancet 1992;339:441–5. (7414):557–60. Prandoni 2002 Hirsh 1993 Bernardi E, Simioni P, Girolami B, Marchiori A, Sabbion Hirsh J. Low molecular weight heparin. Thrombosis & P, Prins M, et al.Recurrent venous thromboembolism and Haemostasis 1993;70(1):204–7. bleeding complications during anticoagulant treatment in patients with cancer and venous thrombosis. Blood 2002; Hirsh 2008 100(10):3484–8. Hirsh J, Bauer KA, Donati MB, Gould M, Samama MM, RevMan 2008 Weitz JI. Parenteral anticoagulants: American College The Nordic Cochrane Centre, The Cochrane Collaboration. of Chest Physicians Evidence-Based Clinical Practice Review Manager (RevMan). 5.0. Copenhagen: The Nordic Guidelines (8th Edition). Chest 2008;133(Suppl):141s–59s. Cochrane Centre, The Cochrane Collaboration, 2008. Karnofsky 1948 Salat 1990 Karnofsky D. The use of nitrogen mustard in the palliative Salat C, Breitruck H, Reinhardt B, Hiller E. treatment of cancer. Cancer 1948;1:634–56. Thromboprophylaxis with low molecular weight heparin (LMWH) and conventional low dose heparin in patients Kuderer 2007 with malignant diseases. Blut 1990;61(2-3):142. Kuderer NM, Khorana AA, Lyman GH, Francis CW. A meta-analysis and systematic review of the efficacy and Smorenburg 1999 safety of anticoagulants as cancer treatment: Impact on Smorenburg SM, Hettiarachchi RJ, Vink R, Buller HR. survival and bleeding complications. Cancer 2007;110(5): The effects of unfractionated heparin on survival in patients 1149–61. with malignancy - a systematic review. Thrombosis and Haemostasis 1999;82(6):1600–4. Lazo-Langner 2007 Smorenburg 2001 Lazo-Langner A, Goss GD, Spaans JN, Rodger MA. The Smorenburg SM, Van Noorden CJ. The complex effects effect of low-molecular-weight heparin on cancer survival. of heparins on cancer progression and metastasis in A systematic review and meta-analysis of randomized trials. experimental studies. Pharmacological Reviews 2001;53(1): Journal of Thrombosis and Haemostasis 2007;5(4):729–37. 93–105. Levine 2003 Thodiyil 2002 Levine MN, Lee AY, Kakkar AK. From Trousseau to targeted Thodiyil P, Kakkar AK. Can low-molecular-weight heparins therapy: new insights and innovations in thrombosis and improve outcomes in patients with cancer?. Cancer cancer. Journal of Thrombosis and Haemostasis 2003;1(7): Treatment Reviews 2002;28:151–5. 1456–63. ∗ Indicates the major publication for the study

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 26 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. CHARACTERISTICSOFSTUDIES

Characteristics of included studies [ordered by study ID]

Agnelli 2009

Methods Randomized clinical trial

Participants 1150 patients with metastatic or locally advanced lung, breast, gastrointestinal (stomach, colon-rectum, pancreas), ovarian or head and neck cancer undergoing chemotherapy

Interventions Subcutaneous low molecular weight heparin (nadroparin calcium, 1 injection/day) versus placebo for the overall duration of chemotherapy or up to a maximum of 4 months

Outcomes • Survival (4 months and 12 months follow up) • Response to chemotherapy (4 months follow up) • For patients with central venous catheters (CVC), complications of possible thrombotic origin, such as malfunction or requirement of CVC removal (4 months follow up) • Superficial thrombophlebitis of lower limbs (4 months follow up) • Asymptomatic thromboembolic events diagnosed during tests performed for other purposes (4 months follow up) • Safety (major bleeding, minor bleeding, other adverse events) (4 months follow up)

Notes Funding: Italfarmaco Follow up: median duration of 111 and 113 days in the nadroparin and placebo groups, respectively NCT 00951574

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Quote: “The randomisation list was generated by an inde- pendent statistician who used a standard permuted block of six without stratification. The list was generated with SAS version 8.2.”

Allocation concealment? Yes Quote: “the system assigned the next free number in accor- dance with the randomisation sequence. Patients and inves- tigators did not know whether study drug or placebo was being given, since pre-filled syringes were used which were identical in appearance.” Comment: definitely yes

Blinding of patients? Yes Quote: “Treatment assignments were masked from all study All outcomes personnel and participants for the duration of the study.” Comment: definitely yes

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 27 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Agnelli 2009 (Continued)

Blinding of providers? Yes Quote: “Treatment assignments were masked from all study All outcomes personnel and participants for the duration of the study.” Comment: definitely yes

Blinding of data collectors? Yes Quote: “Treatment assignments were masked from all study All outcomes personnel and participants for the duration of the study.” Comment: definitely yes

Blinding of outcome adjudicators? Yes Quote: “all the study outcomes were assessed by an inde- All outcomes pendent adjudication committee, whose members were un- aware of the patients’ study group allocation.” Comment: definitely yes

Blinding of data analysts? No Quote: “For the final analysis, the treatment code was All outcomes opened after the database was locked.” Comment: definitely no

Incomplete outcome data addressed? Yes 98.5% followed up All outcomes

Free of selective reporting? Yes As compared to ClinicalTrials.gov information All outcomes

Free of other bias? Yes Study not reported as stopped early for benefit All outcomes

Intention-to-treat analysis? Yes Quote: “All randomised patients who received at least one All outcomes dose of study treatment were included in the efficacy and safety analyses”

Altinbas 2004

Methods Randomized clinical trial

Participants 84 patients with histologically confirmed small cell lung cancer (both limited and exten- sive) undergoing combination chemotherapy, ECOG performance status < 3; median age 58; 81% males; country: Turkey

Interventions LMWH (Dalteparin; prophylactic dose) versus no intervention for 18 weeks or less if disease progressed

Outcomes Outcomes: all-cause mortality (at 12 and 24 months), symptomatic DVT (no PE events; personal communication with author), and bleeding. Screening and diagnostic testing for DVT: not reported

Notes Funding: not reported Follow up: median of 10 months; range 2 to 33 months HR not adjusted (analyses were univariate)

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 28 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Altinbas 2004 (Continued)

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Quote: “open list generated by computer program” (per- sonal communication with author) Comment: definitely yes

Allocation concealment? No Quote: “open list generated by computer program” (per- sonal communication with author) Comment: definitely no

Blinding of patients? No Personal communication with author All outcomes

Blinding of providers? No Personal communication with author All outcomes

Blinding of data collectors? No Personal communication with author All outcomes

Blinding of outcome adjudicators? No Only for staging and evaluation of response to the treatment All outcomes but not for the outcomes of interest (personal communica- tion with authors)

Blinding of data analysts? No Personal communication with author All outcomes

Incomplete outcome data addressed? Yes Complete follow up (personal communication with author) All outcomes

Free of selective reporting? Yes Study not registered. No published protocol. No outcomes All outcomes listed in the methods section. However, all outcomes of in- terest, except for quality of life, reported.

Free of other bias? Yes Study not reported as stopped early for benefit All outcomes

Intention-to-treat analysis? Yes Quote: “Survival was analyzed on an intent-to-treat basis” All outcomes

Kakkar 2004

Methods Randomized clinical trial

Participants 385 patients with histologically confirmed advanced stage III or IV (locally advanced or metastatic) malignant disease of the breast, lung, gastrointestinal tract, pancreas, liver, genitourinary tract, ovary or uterus; minimum life expectancy 3 months; median age 61 IQR (53-68), 43% males; countries: United Kingdom and Canada

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 29 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Kakkar 2004 (Continued)

Interventions LMWH (Dalteparin; prophylactic dose) versus placebo for 12 months; self-injected on a daily basis; no restriction on concomitant chemotherapy or radiotherapy

Outcomes Outcomes: mortality (at 12, 24 and 36 months), symptomatic venous thromboembolism (PE, DVT), major bleeding and minor bleeding Screening testing for DVT/PE: none Diagnostic testing for DVT/PE: “diagnosis determined according to local practices”; not reviewed centrally

Notes Funding: Pharmacia Corp, NY Follow up: maximum of 77 months

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Quote: “Randomization was performed centrally by com- puter-generated code”

Allocation concealment? Yes Quote: “Randomization was performed centrally by com- puter-generated code” Comment: central randomization

Blinding of patients? Yes Quote: “double-blind, placebo controlled study” All outcomes Comment: probably yes

Blinding of providers? Yes Quote: “double-blind, placebo controlled study” All outcomes Comment: probably yes

Blinding of data collectors? Yes Quote: “double-blind, placebo controlled study” All outcomes Comment: probably yes

Blinding of outcome adjudicators? Yes Quote: “double-blind, placebo controlled study” All outcomes Comment: probably yes

Blinding of data analysts? No Quote: “double-blind, placebo controlled study” All outcomes Comment: probably no

Incomplete outcome data addressed? Yes Quote: “Withdrawal of consent before commencing the All outcomes study medication resulted in 11 patients not being included in the analyses. The remaining 374 patients were analyzed for both efficacy and safety” Comment: 100% follow up

Free of selective reporting? Yes Study not registered. No published protocol. All outcomes All outcomes listed in methods section are reported on in the results sec- tion. All outcomes of interest, except for quality of life, re-

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 30 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Kakkar 2004 (Continued)

ported.

Free of other bias? Yes Study not reported as stopped early for benefit All outcomes

Intention-to-treat analysis? Yes Quote: “All patients who gave informed consent and who All outcomes had at least one injection of study drug or placebo consti- tuted the intent-to-treat population for efficacy and safety analyses” Comment: investigators excluded patients who did not have at least one injection of study drug or placebo

Klerk 2005

Methods Randomized clinical trial

Participants 302 patients with different types of solid malignant tumors, “that could not be treated curatively” including: colorectal, breast, lung gastric, oesophageal, liver, gallbladder, Katskin, prostate, pancreatic, cervical, urothelial, renal, ovarian, melanoma, endome- trial and other cancers; minimum life expectancy 1 month, stratified according to life expectancy (< or > 6 months); median age 64; 52% males

Interventions LMWH (Nadroparin) versus placebo for 6 weeks; 2 weeks therapeutic dose then 4 weeks prophylactic dose; concomitant antineoplastic therapy

Outcomes Outcomes: mortality from any cause (at 6, 12 and 24 months), major bleeding, clini- cally relevant non-major bleeding, all clinically relevant bleeding (major and non-major combined)

Notes Funding: Sanofi provided study medication Follow up: mean of 12 months HR adjusted for: life expectancy (< 6 versus >= 6 months), WHO performance status (1 or less, 2, 3 or more) concomitant treatment (chemotherapy, radiotherapy, hormonal therapy, other antineoplastic treatment), type of cancer (breast, colorectal, cervical or other) and histology (adeno, squamous, other)

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Quote: “Sequentially numbered boxes of syringes with nadroparin or placebo were prepared using a central com- puter-generated randomization schedule”

Allocation concealment? Yes Quote: “Sequentially numbered boxes of syringes with nadroparin or placebo were prepared using a central com- puter-generated randomization schedule” Comment: central randomization

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 31 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Klerk 2005 (Continued)

Blinding of patients? Yes Quote: “double-blind, placebo controlled study”. Per- All outcomes sonal communication with authors: “Patients, healthcare providers, data collectors and outcome adjudicators were blinded.” Comment: definitely yes

Blinding of providers? Yes Quote: “double-blind, placebo controlled study”. Per- All outcomes sonal communication with authors: “Patients, healthcare providers, data collectors and outcome adjudicators were blinded.” Comment: definitely yes

Blinding of data collectors? Yes Quote: “double-blind, placebo controlled study”. Per- All outcomes sonal communication with authors: “Patients, healthcare providers, data collectors and outcome adjudicators were blinded.” Comment: definitely yes

Blinding of outcome adjudicators? Yes Quote: “double-blind, placebo controlled study”. Per- All outcomes sonal communication with authors: “Patients, healthcare providers, data collectors and outcome adjudicators were blinded.” Comment: definitely yes

Blinding of data analysts? No Quote: “double-blind, placebo controlled study”. Personal All outcomes communication with authors: “After the study had ended the data analysts had access to the unblinded data.” Comment: definitely no

Incomplete outcome data addressed? Yes Quote: “All patients were observed until death or until the All outcomes end of the study”. “No patients were lost to follow-up” Comment: 100% follow up

Free of selective reporting? Yes Study not registered. No published protocol. All outcomes All outcomes listed in methods section are reported on in the results sec- tion. However study does not report number of thrombotic events. Personal communication with authors: “VTE was not an endpoint of the study and it was not standardly reg- istered per protocol”.

Free of other bias? Yes Study not reported as stopped early for benefit All outcomes

Intention-to-treat analysis? Yes Quote: “All primary analyses were performed on an inten- All outcomes tion-to-treat principle”

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 32 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Lebeau 1994

Methods Randomized clinical trial

Participants 277 patients with histologically diagnosed small cell lung cancer both limited and ex- tensive; 78% had Karnofsky > 80; 85% older than 50; 91% males

Interventions UFH (prophylactic dose) versus no intervention for 5 weeks; 2 or 3 daily subcutaneous injections; in combination with chemotherapy

Outcomes Outcomes: overall survival (at 12, 24 and 36 months), bleeding

Notes Funding: none Follow up: maximum of 84 months

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Quote: “Randomized through a centralized blind telephone assignment procedure”

Allocation concealment? Yes Quote: “Randomized through a centralized blind telephone assignment procedure” Comment: central randomization

Blinding of patients? No Quote: “no blinding procedure for patients and physicians All outcomes was used because overall survival was chosen as the major endpoint” Comment: definitely no

Blinding of providers? No Quote: “no blinding procedure for patients and physicians All outcomes was used because overall survival was chosen as the major endpoint” Comment: definitely no

Blinding of data collectors? No Quote: “no blinding procedure for patients and physicians All outcomes was used because overall survival was chosen as the major endpoint” Comment: definitely no

Blinding of outcome adjudicators? No Quote: “no blinding procedure for patients and physicians All outcomes was used because overall survival was chosen as the major endpoint” Comment: definitely no

Blinding of data analysts? No Quote: “no blinding procedure for patients and physicians All outcomes was used because overall survival was chosen as the major endpoint” Comment: definitely no

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 33 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Lebeau 1994 (Continued)

Incomplete outcome data addressed? Yes Quote: “No patient was lost to follow-up” All outcomes Comment: 100% follow up

Free of selective reporting? Unclear Study not registered. No published protocol. The outcomes All outcomes listed in methods section are reported on in the results sec- tion. However study does not report number of thrombotic events.

Free of other bias? Yes Study not reported as stopped early for benefit All outcomes

Intention-to-treat analysis? Yes Quote: “Analysis was made on an intention-to-treat basis” All outcomes

Pelzer 2009

Methods Open, prospective, randomized, multicenter phase III study

Participants 312 chemotherapy-naive patients with histologically or cytologically confirmed advanced pancreatic cancer

Interventions LMWH (enoxaparin intermediate dose - 1 mg/kg daily for the first 3 months followed by 40 mg daily an additional 3 months) versus no intervention; simultaneous initiation of palliative systemic chemotherapy

Outcomes Overall survival, symptomatic VTE, major bleeding, time to progression (TTP)

Notes Median follow up of 30.4 weeks CCT-NAPN-16752, ISRCTN02140505 Funding: Forschungsförderung der Charité, Deutsche Krebshilfe, Lilly, Amgen, Sanofi- aventis

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Quote: “computer generated, dynamic, central randomisation” (personal commu- nication with author) Comment: definitely yes

Allocation concealment? Yes Quote: “computer generated, dynamic, central randomisation” (personal commu- nication with author) Comment: definitely yes

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 34 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Pelzer 2009 (Continued)

Blinding of patients? No Quote: “Open, prospective, randomized, All outcomes multicenter phase III study” Comment: definitely no

Blinding of providers? No Quote: “Open, prospective, randomized, All outcomes multicenter phase III study” Comment: definitely no

Blinding of data collectors? No Quote: “Open, prospective, randomized, All outcomes multicenter phase III study” Comment: definitely no

Blinding of outcome adjudicators? No Quote: “Open, prospective, randomized, All outcomes multicenter phase III study” Comment: definitely no

Blinding of data analysts? Yes Quote: “data analysts were blinded for che- All outcomes motherapy design and for enoxaparin treat- ment” (personal communication with au- thor) Comment: definitely yes

Incomplete outcome data addressed? Yes 94% follow up for thromboembolic events All outcomes and 87% follow up for survival (personal communication with author)

Free of selective reporting? Unclear Awaiting full publication. Mortality and All outcomes minor bleeding data not available as of May 2010. Awaiting full publication

Free of other bias? Yes Study not reported as stopped early for ben- All outcomes efit. However, 312 patients were recruited into the trial whereas the earlier published abstracts (Pelzer 2005; Pelzer 2007) re- ported a target recruitment of 540 patients. Awaiting full publication

Intention-to-treat analysis? Yes Quote: “ITT and PP analysis” All outcomes

Perry 2010

Methods Randomized, placebo-controlled trial

Participants 186 adults with newly diagnosed malignant glioma

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 35 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Perry 2010 (Continued)

Interventions LMWH dalteparin sodium 5000 International Units or saline placebo, both adminis- tered subcutaneously once daily. The duration of therapy for the primary analysis of efficacy and safety was 6 months. The intention was for patients to continue study med- ication (dalteparin or placebo) for up to an additional 6 months.

Outcomes Primary outcome was objectively documented symptomatic DVT or pulmonary em- bolism. Secondary outcomes were bleeding (major and all bleeding), quality of life, cog- nition assessments and death. The duration of follow up was 12 months.

Notes ClinicalTrials.gov Identifier: NCT00135876 Funding: Pfizer Inc, Ontario Clinical Oncology Group, Crolla Chair in Brain Tumour Research (JP) HR adjusted for the following strata: center, tumor grade (3 versus 4), the Karnofsky Performance Status (KPS) (< 60 versus 70 or more), and time from surgery to random- ization (< 2 weeks versus 2 to 4 weeks) We noted the following discrepancy: in the abstract, the 12-month mortality rates were reported as 47.8% for LMWH and 45.4% for placebo (which correspond to 47 and 40 events); in the text the number of events are reported as being 45 and 32

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Quote: “Treatment allocations were pre-deter- mined using a computer-generated randomiza- tion list with random size permuted blocks, and stratified according to center, tumor grade (3 versus 4), the Karnofsky Performance Status (KPS) (< 60 versus 70 or more), and time from surgery to randomization (< 2 weeks versus 2- 4 weeks).” Comment: definitely yes

Allocation concealment? Yes Quote: “Consenting patients were random- ized by contacting the Ontario Clinical Oncol- ogy Group (OCOG) Coordinating and Meth- ods Centre at the Henderson Research Centre, Hamilton, Ontario.” Comment: definitely yes

Blinding of patients? Yes Quote: “investigators, patients and outcome as- All outcomes sessors were blinded to treatment allocation.” Comment: definitely yes

Blinding of providers? Yes Quote: “investigators, patients and outcome as- All outcomes sessors were blinded to treatment allocation.” Comment: probably yes

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 36 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Perry 2010 (Continued)

Blinding of data collectors? Yes Quote: “investigators, patients and outcome as- All outcomes sessors were blinded to treatment allocation.” Comment: probably yes

Blinding of outcome adjudicators? Yes Quote: “investigators, patients and outcome as- All outcomes sessors were blinded to treatment allocation.” Comment: definitely yes

Blinding of data analysts? No Quote: “investigators, patients and outcome as- All outcomes sessors were blinded to treatment allocation.” Comment: probably no

Incomplete outcome data addressed? Yes Although not clearly stated, 100% of patients All outcomes randomized were analyzed Comment: probably yes

Free of selective reporting? Yes Based on trial registration and methods section: All outcomes yes for death, VTE, bleeding; no for quality of life and cognition assessments

Free of other bias? Yes Stopped early but for slow accrual All outcomes

Intention-to-treat analysis? Yes Quote: “The intention-to-treat principle was All outcomes used in all analyses.”

Sideras 2006

Methods Blinded, placebo-controlled, randomized clinical trial changed to open-labelled clinical trial

Participants 141 patients with advanced breast, prostate, lung or colorectal cancer with minimum life expectancy 12 weeks; ECOG performance status 0-2

Interventions LMWH (Dalteparin; prophylactic dose) versus placebo for unclear duration; then changed to LMWH (Dalteparin; prophylactic dose) vs no intervention; duration not specified; with concomitant chemotherapy and/or radiotherapy

Outcomes Outcomes: overall survival (at 12, 24 and 36 months), VTE and bleeding, quality of life (measured by Uniscale, and the Symptom Distress Scale (SDS)) Screening testing for DVT/PE: none Diagnostic testing for DVT/PE: decided by the primary clinician

Notes Funding: governmentally funded, pharmaceutical company supplied drug and placebo Follow up: maximum of 24 months HR adjusted for: treatment arm; baseline good, bad, or unsure status; tumor site; and baseline performance score

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 37 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Sideras 2006 (Continued)

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Quote: “Randomization handled through the North Central Cancer Treatment Group (NCGTG) Randomization Office using a dynamic allocation method”

Allocation concealment? Yes Quote: “Randomization handled through the North Central Cancer Treatment Group (NCGTG) Randomization Office using a dynamic allocation method” Comment: central randomization

Blinding of patients? No Comments: All outcomes • Initially, the study was double- blinded and placebo-controlled. However, because of low accrual, the study became open-labelled. • probably yes initially; then definitely no

Blinding of providers? No Comments: All outcomes • Initially, the study was double- blinded and placebo-controlled. However, because of low accrual, the study became open-labelled. • probably yes initially; then definitely no

Blinding of data collectors? No Comments: All outcomes • Initially, the study was double- blinded and placebo-controlled. However, because of low accrual, the study became open-labelled. • probably yes initially; then definitely no

Blinding of outcome adjudicators? No Comments: All outcomes • Initially, the study was double- blinded and placebo-controlled. However, because of low accrual, the study became open-labelled. • probably yes initially; then definitely no

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 38 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Sideras 2006 (Continued)

Blinding of data analysts? No Comments: All outcomes • Initially, the study was double- blinded and placebo-controlled. However, because of low accrual, the study became open-labelled. • probably yes initially; then definitely no

Incomplete outcome data addressed? Yes Comment: 98% follow up All outcomes

Free of selective reporting? Yes All outcomes of interest reported All outcomes

Free of other bias? Yes Quote: “The protocol accrual was stopped All outcomes before reaching the prestudy planned ac- crual goal by the NCCTG Data Monitor- ing Committee because of a slower than predicted protocol accrual rate, with the knowledge (provided by an interim anal- ysis report) that the patient survival rates were numerically worse on one arm of the blinded study.” Comment: study was stopped early for in- sufficient accrual but not for benefit

Intention-to-treat analysis? Unclear Not reported All outcomes

Weber 2008

Methods Prospective, open, randomized study

Participants 20 patients aged 55 to 88 years with advanced cancer (19 solid cancer and 1 hematological cancer) with a minimum life expectancy of 6 months; 45% males

Interventions LMWH (Nadroparin, prophylactic dose) versus no intervention administered subcuta- neously on a daily basis for unclear duration; with concomitant anticancer treatment

Outcomes Outcomes: mortality (at 3, 6, 12 and 15 months), symptomatic venous thromboem- bolism (PE, DVT), major bleeding, minor bleeding and thrombocytopenia Screening testing for DVT/PE: none Diagnostic testing for DVT/PE: echo-doppler for DVT and spiral CT scan for PE

Notes Funding: not reported Follow up: maximum of 15 months

Risk of bias

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 39 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Weber 2008 (Continued)

Item Authors’ judgement Description

Adequate sequence generation? Yes Quote: “The sequence of treatments was ran- domly assigned in blocks of constant size (n = 20).” Comment: definitely yes

Allocation concealment? Yes Quote: “Sets of 20 sealed envelopes (10 Yes and 10 No) were numbered consecutively. The se- quence of treatments was randomly assigned in blocks of constant size.” Comment: probably yes

Blinding of patients? No Quote: “Prospective open randomized study” All outcomes Comment: definitely no

Blinding of providers? No Quote: “Prospective open randomized study” All outcomes Comment: definitely no

Blinding of data collectors? No Quote: “Prospective open randomized study” All outcomes Comment: definitely no

Blinding of outcome adjudicators? No Quote: “Prospective open randomized study” All outcomes Comment: definitely no

Blinding of data analysts? No Quote: “Prospective open randomized study” All outcomes Comment: probably no

Incomplete outcome data addressed? Yes Quote: “No patient was lost to follow-up” All outcomes Comment: 100% follow up

Free of selective reporting? Yes Study not registered. No published protocol. All outcomes The outcomes listed in methods section are re- ported on in the results section.

Free of other bias? Yes Study not reported as stopped early for benefit All outcomes

Intention-to-treat analysis? Yes Quote: “Excluded from the analysis (n = 0)” All outcomes Comment: all patients randomized to treatment or control group were included in the analysis

CT: computerised tomography; CVC: central venous catheter; DVT: deep vein thrombosis; ECOG: Eastern Co-operative Oncology Group; HR: hazard ratio; ITT: intention-to-treat; LMWH: low molecular weight heparin; PE: pulmonary embolism; PP: per protocol; TTP: time to progression

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 40 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Characteristics of excluded studies [ordered by study ID]

Study Reason for exclusion

Alifano 2005 Letter to the editor

Arbit 2005 Letter to the editor

Barberi-Heyob 1995 Letter to the editor

Barkagan 1997 Not the comparison of interest (LMWH versus vitamin K antagonists versus UFH)

Bitsch 1990 Topical heparin

Blaszczyk 1970 Not randomized

Buckman 2005 No control group

Craven 2001 Letter to editor

Crossno 2009 Letter to editor

Di Nisio 2005 Review

Edlis 1976 No control group

Elias 1972 Case series

Elias 1973a Not randomized

Elias 1973b Case series

Elias 1973c Not randomized

Elias 1974 Case series

Elias 1975 Not randomized

Fielding 1992 Intraportal infusion with heparin

Graf 1994 Not the comparison of interest (LMWH versus vitamin K antagonists)

Graf 1996 Not the comparison of interest (LMWH versus vitamin K antagonists)

Green 1992 Letter to the editor

Guimbretiere 1982 Not randomized

Jorgensen 2001 No control group

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 41 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. (Continued)

Kohanna 1983 Retrospective study

Lecumberri 2005 Review

Lemoine 2005 Editorial

Levine 2005 Treatment for DVT/PE; letter

Loynes 2002 Case report

Lykke 2003 Review

Mammen 2004 Preface

Meyer 2007 Letter to editor

Mousa 2001 No cancer patients in the study

Munstedt 1996 Intervention consisted of only 2 doses of LMWH

Nash 2000 Letter to editor

Nishioka 2007 Review

Nitti 1997 Intraportal infusion with heparin

Retik 1962 No cancer patients in the study

Rohwedder 1977 No control group

Siragusa 1999 Letter to the editor

Spigel 2005 Review

Von Hugo 1981 No survival outcome

Wojtukiewicz 2003 No control group

Zacharski 2003 Editorial

DVT: deep vein thrombosis; LMWH: low molecular weight heparin; PE: pulmonary embolism; UFH: unfractionated heparin

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 42 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Characteristics of ongoing studies [ordered by study ID]

ABEL

Trial name or title Adjuvant BEmiparin in Small Cell Lung Carcinoma (ABEL STUDY)

Methods Study type: interventional Allocation: randomized Control: active control Endpoint classification: safety/efficacy study Intervention model: parallel assignment Masking: open label

Participants Patients diagnosed with limited small cell lung cancer

Interventions Subcutaneous administration (sc) of Bemiparin (3500 UI/day) for 26 weeks, starting on the first day of chemotherapy (CT) versus no Bemiparin

Outcomes • efficacy: progression-free survival in months (measurements of the size of the effect will be done using the Kaplan Meier method and the difference between the means of this time) (time frame: efficacy) • safety: will be the incidence, during randomized treatment period (from day 1 to the last day of treatment + 7 days), of major bleedings (time frame: safety)

Starting date June 2005

Contact information Eduardo Rocha, MD (+34) 948 25 54 00 [email protected]

Notes Study Director: Eduardo Rocha, MD; Clínica Universitaria de Navarra Sponsor: Instituto Cientifico y Tecnologico de Navarra, Universidad de Navarra This study is recruiting participants as of May 2010 http://clinicaltrials.gov/ct2/show/NCT00324558

FOCUS

Trial name or title A Phase II Randomized Study of Fragmin in Ovarian Cancer: Utility on Survival (FOCUS)

Methods Study Type: Interventional Allocation: Randomized Control: Dose Comparison Endpoint Classification: Safety/Efficacy Study Intervention Model: Parallel Assignment Masking: Open Label

Participants Patients with newly diagnosed epithelial ovarian carcinoma

Interventions One of 3 doses of a LMWH dalteparin in conjunction with standard adjuvant taxane and platinum-based chemotherapy: 50, 100, 150 IU/kg administered subcutaneously once daily for 3 cycles of chemotherapy

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 43 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. FOCUS (Continued)

Outcomes Primary outcome: disease response Secondary outcomes: • symptomatic venous thromboembolism (time frame: up to 7 days after last dose of dalteparin) • bleeding (time frame: up to 24 hours after last dose of dalteparin) • compliance (time frame: up to the end of cycle 3) • death (time frame: up to the last day of follow up)

Starting date October 2005

Contact information Mark Levine, MD McMaster University, Ontario Clinical Oncology Group

Notes Principal Investigator: Mark Levine, MD McMaster University, Ontario Clinical Oncology Group Study sponsor: Ontario Clinical Oncology Group (OCOG) This study was completed in January 2010 http://clinicaltrials.gov/ct2/show/NCT00239980

FRAGMATIC

Trial name or title FRAGMATIC - A Randomised Phase III Clinical Trial Investigating the Effect of FRAGMin Added to Standard Therapy In Patients With Lung Cancer

Methods Study type: interventional Allocation: randomized Control: active control Masking: open-label

Participants 2200 patients with histological or cytological diagnosis of primary bronchial carcinoma (small cell or non- small cell) within the last weeks; ECOG performance status 0, 1, 2 or 3; willing to receive LMWH

Interventions • Arm I (control): patients receive anticancer treatment considered appropriate by the local medical team • Arm II: patients receive anticancer treatment considered appropriate by the local medical team. Beginning before the start of the first definitive anticancer treatment, patients receive dalteparin subcutaneously daily for up to 24 weeks.

Outcomes Primary outcome measures: • Overall survival Secondary outcome measures: • Venous thrombotic event-free survival • Serious adverse events • Metastasis-free survival • Toxicity • Quality of life as measured by EQ-5D • Breathlessness (dyspnea) as measured by the Cancer Dyspnea Scale • Anxiety and depression as measured by the Hospital Anxiety and Depression Scale • Cost effectiveness • Cost utility Quality of life, anxiety, depression and dyspnea are assessed at baseline, at 12 and 24 weeks, and then at 9

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 44 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. FRAGMATIC (Continued)

and 12 months After completion of therapy, patients are followed at 9 months, 1 year and then every 6 months for at least 2 years

Starting date August 2007

Contact information griffi[email protected]

Notes Study Chair: Fergus Macbeth, MD; Velindre NHS Trust Sponsors and Collaborators: Velindre NHS Trust This study is recruiting participants as of May 2010 http://clinicaltrials.gov/ct2/show/NCT00519805

GASTRANOX

Trial name or title Overall Survival of Inoperable Gastric/GastroOesophageal Cancer Subjects on Treating With LMWH + Chemotherapy(CT) vs Standard CT (GASTRANOX)

Methods Study type: interventional Allocation: randomized Control: active control Endpoint classification: efficacy study Intervention model: parallel assignment Masking: open-label

Participants Patients with inoperable gastric and gastro-oesophageal cancer

Interventions Enoxaparin (once daily dose of 1 mg/kg of body weight for 6 months) given concomitantly with chemotherapy vs chemotherapy alone

Outcomes Primary outcome measures: event-free survival (EFS) - composite endpoint of overall survival plus free of symptomatic VTE (time frame: up to 1 year from start of treatment) Secondary outcome measures: incidence of symptomatic VTE, overall survival, major and minor hemorrhages during chemotherapy and/or up to 30 days after last dose is provided. Serious adverse events, all reported adverse events, heparin induced thrombocytopenia (time frame: up to 1 year from the start of treatment)

Starting date July 2008

Contact information Janice M Maganji, MBBS. 00 44 7824836535; [email protected]

Notes Principal Investigator: Ajay K Kakkar, PhD Thrombosis Research Institute Sponsors and Collaborators Thrombosis Research Institute This study is recruiting participants as of May 2010 http://clinicaltrials.gov/ct2/show/NCT00718354

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 45 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. INPACT

Trial name or title The INPACT Study (Improving With Nadroparin the Prognosis in Advanced Cancer Treatment). A Ran- domized, Controlled Trial to Evaluate the Effects of Nadroparin on Survival and Disease Progression in Pa- tients With Advanced Malignancies of the Lung, Pancreas, or Prostate

Methods Study type: interventional Allocation: randomized Endpoint classification: safety/efficacy study Intervention model: parallel assignment Masking: open label

Participants Hormone refractory prostate cancer within 6 months of diagnosis, OR locally advanced (non-metastasized) pancreatic cancer within 3 months of diagnosis, OR non-small cell lung cancer within 3 months of stage IIIB

Interventions Nadroparin versus no nadroparin. All received standard anticancer treatment. Nadroparin patients treated with therapeutic doses of subcutaneous nadroparin for 2 weeks followed by half therapeutic doses for 4 weeks. After 4 weeks of wash-out, subsequent 2-week periods of therapeutic doses of nadroparin will be given for a total of 6 cycles each separated by a 4-week wash-out.

Outcomes Primary outcome measures: death due to all causes at study end (patients will be followed until at least week 46 after randomization) (time frame: at least 46 weeks after randomization) Secondary outcome measures: time to tumor progression (time frame: 46 weeks)

Starting date May 2006

Contact information -

Notes Study Director: GSK Clinical Trials; GlaxoSmithKline Sponsor: GlaxoSmithKline This study was completed in July 2009. Preliminary results presented at XXII Congress of the ISTH. http://clinicaltrials.gov/ct2/show/NCT00312013

NCT00916669

Trial name or title A Randomized Phase II Study to Evaluate the Effect of Two Different Doses of Enoxaparin Sodium in Combination With Standard Chemotherapy (Cisplatin Plus Etoposide) With Respect to Time to Tumor Progression (TTP) in Patients With Newly Diagnosed Extensive Stage Small Cell Lung Cancer (SCLC) Without Underlying Venous Thromboembolism

Methods Study type: interventional Allocation: randomized Endpoint classification: safety/efficacy Study Intervention model: parallel assignment Masking: open-label

Participants Patients with newly diagnosed extensive stage SCLC without underlying venous thromboembolism

Interventions Group A: active comparator; cisplatin and etoposide Group B: experimental; cisplatin and etoposide, plus low-dose enoxaparin sodium

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 46 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. NCT00916669 (Continued)

Group C: experimental; cisplatin and etoposide, plus high-dose enoxaparin sodium

Outcomes Primary outcome measures: to evaluate the prophylactic and treatment doses of enoxaparin sodium given in combination with standard chemotherapy compared to standard chemotherapy alone with respect to time to tumor progression in this patient population (time frame: 2 years) Secondary outcome measures: to determine the effect of 2 different doses of enoxaparin sodium in combination with chemotherapy and chemotherapy alone on biomarkers of angiogenesis and to identify if these markers correlate with overall survival and progression-free survival (time frame: 2 years) (designated as safety issue: No) To evaluate toxicity and determine the rates of bleeding complications in this patient population (time frame: 2 years)

Starting date July 2008

Contact information Rachel Rosovsky, MD, MPH

Notes Principal Investigator: Rachel Rosovsky, MD, MPH; Massachusetts General Hospital Sponsors and Collaborators: Massachusetts General Hospital; Dana-Farber Cancer Institute; Beth Israel Deaconess Medical Center; North Shore Medical Center; Sanofi-Aventis This study is ongoing, but not recruiting participants as of May 2010 http://clinicaltrials.gov/ct2/show/NCT00916669

CT: computerised tomography; ECOG: Eastern Co-operative Oncology Group; EFS: event-free survival; LMWH: low molecular weight heparin; SCLC: small cell lung cancer

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 47 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. DATA AND ANALYSES

Comparison 1. Heparin vs placebo

No. of No. of Outcome or subgroup title studies participants Statistical method Effect size

1 Mortality at 12 months 8 2530 Risk Ratio (M-H, Random, 95% CI) 0.93 [0.85, 1.02] 2 Mortality at 24 months 5 1174 Risk Ratio (M-H, Random, 95% CI) 0.92 [0.88, 0.97] 3 Mortality over duration of study 7 1381 Hazard Ratio (Random, 95% CI) 0.79 [0.67, 0.93] 4 Symptomatic VTE 7 2264 Risk Ratio (M-H, Random, 95% CI) 0.55 [0.37, 0.82] 5 Major bleeding 9 2843 Risk Ratio (M-H, Random, 95% CI) 1.30 [0.59, 2.88] 6 Minor bleeding 7 2345 Risk Ratio (M-H, Random, 95% CI) 1.05 [0.75, 1.46]

Analysis 1.1. Comparison 1 Heparin vs placebo, Outcome 1 Mortality at 12 months.

Review: Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation

Comparison: 1 Heparin vs placebo

Outcome: 1 Mortality at 12 months

Studyorsubgroup Heparin Control RiskRatio Weight RiskRatio n/N n/N M-H,Random,95% CI M-H,Random,95% CI Agnelli 2009 333/769 155/381 19.7 % 1.06 [ 0.92, 1.23 ]

Perry 2010 45/99 32/87 6.0 % 1.24 [ 0.87, 1.75 ]

Weber 2008 8/10 10/10 6.1 % 0.81 [ 0.57, 1.14 ]

Klerk 2005 88/148 107/154 16.9 % 0.86 [ 0.72, 1.01 ]

Kakkar 2004 103/190 109/184 16.0 % 0.92 [ 0.77, 1.09 ]

Sideras 2006 40/68 41/69 8.6 % 0.99 [ 0.75, 1.31 ]

Altinbas 2004 22/42 30/42 6.1 % 0.73 [ 0.52, 1.04 ]

Lebeau 1994 96/138 110/139 20.6 % 0.88 [ 0.76, 1.01 ] Total (95% CI) 1464 1066 100.0 % 0.93 [ 0.85, 1.02 ] Total events: 735 (Heparin), 594 (Control) Heterogeneity: Tau2 = 0.01; Chi2 = 10.75, df = 7 (P = 0.15); I2 =35% Test for overall effect: Z = 1.57 (P = 0.12)

0.5 0.7 1 1.5 2 Favours heparin Favours control

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 48 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.2. Comparison 1 Heparin vs placebo, Outcome 2 Mortality at 24 months.

Review: Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation

Comparison: 1 Heparin vs placebo

Outcome: 2 Mortality at 24 months

Studyorsubgroup Heparin Control RiskRatio Weight RiskRatio n/N n/N M-H,Random,95% CI M-H,Random,95% CI Altinbas 2004 35/42 42/42 12.3 % 0.84 [ 0.73, 0.96 ]

Lebeau 1994 122/138 128/139 40.6 % 0.96 [ 0.89, 1.04 ]

Sideras 2006 54/68 56/69 8.9 % 0.98 [ 0.83, 1.16 ]

Kakkar 2004 139/190 151/184 20.4 % 0.89 [ 0.80, 0.99 ]

Klerk 2005 111/148 128/154 17.8 % 0.90 [ 0.80, 1.01 ] Total (95% CI) 586 588 100.0 % 0.92 [ 0.88, 0.97 ] Total events: 461 (Heparin), 505 (Control) Heterogeneity: Tau2 = 0.0; Chi2 = 3.95, df = 4 (P = 0.41); I2 =0.0% Test for overall effect: Z = 3.26 (P = 0.0011)

0.5 0.7 1 1.5 2 Favours heparin Favours control

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 49 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.3. Comparison 1 Heparin vs placebo, Outcome 3 Mortality over duration of study.

Review: Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation

Comparison: 1 Heparin vs placebo

Outcome: 3 Mortality over duration of study

Studyorsubgroup Heparin Control log[HazardRatio] Hazard Ratio Weight HazardRatio N N (SE) IV,Random,95% CI IV,Random,95% CI

Altinbas 2004 42 42 -0.6531 (0.2321) 9.7 % 0.52 [ 0.33, 0.82 ]

Kakkar 2004 190 184 -0.2395 (0.1103) 22.4 % 0.79 [ 0.63, 0.98 ]

Klerk 2005 148 154 -0.2838 (0.1123) 22.1 % 0.75 [ 0.60, 0.94 ]

Lebeau 1994 138 139 -0.334 (0.1222) 20.6 % 0.72 [ 0.56, 0.91 ]

Perry 2010 99 87 0.18 (0.263) 8.1 % 1.20 [ 0.72, 2.00 ]

Sideras 2006 68 70 0.1406 (0.1927) 12.6 % 1.15 [ 0.79, 1.68 ]

Weber 2008 10 10 -0.44 (0.37) 4.6 % 0.64 [ 0.31, 1.33 ] Total (95% CI) 100.0 % 0.79 [ 0.67, 0.93 ] Heterogeneity: Tau2 = 0.02; Chi2 = 10.63, df = 6 (P = 0.10); I2 =44% Test for overall effect: Z = 2.79 (P = 0.0053)

0.2 0.5 1 2 5 Favours heparin Favours control

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 50 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.4. Comparison 1 Heparin vs placebo, Outcome 4 Symptomatic VTE.

Review: Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation

Comparison: 1 Heparin vs placebo

Outcome: 4 Symptomatic VTE

Studyorsubgroup Heparin Control RiskRatio Weight RiskRatio n/N n/N M-H,Random,95% CI M-H,Random,95% CI

Altinbas 2004 0/42 1/42 1.6 % 0.33 [ 0.01, 7.96 ]

Agnelli 2009 11/769 11/381 23.4 % 0.50 [ 0.22, 1.13 ]

Perry 2010 11/99 14/87 29.6 % 0.69 [ 0.33, 1.44 ]

Pelzer 2009 8/160 22/152 26.5 % 0.35 [ 0.16, 0.75 ]

Weber 2008 1/10 0/10 1.7 % 3.00 [ 0.14, 65.90 ]

Sideras 2006 4/68 5/70 9.9 % 0.82 [ 0.23, 2.94 ]

Kakkar 2004 3/190 4/184 7.3 % 0.73 [ 0.16, 3.20 ] Total (95% CI) 1338 926 100.0 % 0.55 [ 0.37, 0.82 ] Total events: 38 (Heparin), 57 (Control) Heterogeneity: Tau2 = 0.0; Chi2 = 3.58, df = 6 (P = 0.73); I2 =0.0% Test for overall effect: Z = 2.92 (P = 0.0035)

0.02 0.1 1 10 50 Favours heparin Favours control

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 51 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.5. Comparison 1 Heparin vs placebo, Outcome 5 Major bleeding.

Review: Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation

Comparison: 1 Heparin vs placebo

Outcome: 5 Major bleeding

Studyorsubgroup Heparin Control RiskRatio RiskRatio n/N n/N M-H,Random,95% CI M-H,Random,95% CI

Agnelli 2009 5/769 0/381 5.46 [ 0.30, 98.43 ]

Altinbas 2004 0/42 0/42 0.0 [ 0.0, 0.0 ]

Kakkar 2004 1/190 0/184 2.91 [ 0.12, 70.87 ]

Klerk 2005 5/148 1/154 5.20 [ 0.62, 44.01 ]

Lebeau 1994 1/138 1/139 1.01 [ 0.06, 15.94 ]

Pelzer 2009 10/160 15/152 0.63 [ 0.29, 1.37 ]

Perry 2010 5/99 1/87 4.39 [ 0.52, 36.89 ]

Sideras 2006 2/68 5/70 0.41 [ 0.08, 2.05 ]

Weber 2008 1/10 0/10 3.00 [ 0.14, 65.90 ] Total (95% CI) 1624 1219 1.30 [ 0.59, 2.88 ] Total events: 30 (Heparin), 23 (Control) Heterogeneity: Tau2 = 0.30; Chi2 = 9.19, df = 7 (P = 0.24); I2 =24% Test for overall effect: Z = 0.66 (P = 0.51)

0.01 0.1 1 10 100 Favours heparin Favours control

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 52 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.6. Comparison 1 Heparin vs placebo, Outcome 6 Minor bleeding.

Review: Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation

Comparison: 1 Heparin vs placebo

Outcome: 6 Minor bleeding

Studyorsubgroup Heparin Control RiskRatio Weight RiskRatio n/N n/N M-H,Random,95% CI M-H,Random,95% CI

Agnelli 2009 57/769 30/381 62.3 % 0.94 [ 0.62, 1.44 ]

Altinbas 2004 1/42 0/42 1.1 % 3.00 [ 0.13, 71.61 ]

Kakkar 2004 8/190 5/184 9.3 % 1.55 [ 0.52, 4.65 ]

Klerk 2005 5/148 0/154 1.3 % 11.44 [ 0.64, 205.14 ]

Lebeau 1994 1/138 0/139 1.1 % 3.02 [ 0.12, 73.54 ]

Sideras 2006 13/68 13/70 23.4 % 1.03 [ 0.52, 2.06 ]

Weber 2008 0/10 2/10 1.3 % 0.20 [ 0.01, 3.70 ] Total (95% CI) 1365 980 100.0 % 1.05 [ 0.75, 1.46 ] Total events: 85 (Heparin), 50 (Control) Heterogeneity: Tau2 = 0.0; Chi2 = 5.52, df = 6 (P = 0.48); I2 =0.0% Test for overall effect: Z = 0.27 (P = 0.79)

0.005 0.1 1 10 200 Favours heparin Favours control

APPENDICES

Appendix 1. Full search strategies for the electronic databases

Database Strategy

MEDLINE #1 Heparin/ #2 Heparin.tw #3 Heparin, Low-Molecular-Weight/ #4 (LMWH OR low molecular weight heparin OR nadroparin OR fraxiparin OR enoxaparin OR clexane OR lovenox OR dalteparin OR fragmin OR ardeparin OR normiflo OR tinzaparin OR logi- parin OR innohep OR certoparin OR sandoparin OR reviparin OR clivarin OR danaproid OR orgaran).tw #51OR2OR3OR4 #6 Coumarins/

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 53 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. (Continued)

#7 Warfarin/ #8 (warfarin OR coumadin OR acenocumarol OR phenprocumon OR 4-hydroxicoumarins OR oral anticoagulant OR vitamin K an- tagonist OR VKA).tw #96OR7OR8 #10 (fondaparinux OR Arixtra).tw #11 (ximelagatran OR Exanta).tw #12 (Pradaxa or Dabigatran or rivaroxaban or Xarelto or apixaban) .tw. #135OR9OR10OR11OR12 #14 Neoplasms/ #15 (malignan$ OR neoplasm$ OR cancer OR carcinoma$ OR adenocarcinoma OR tumour OR tumor).tw #16 14 OR 15 #17 clinical trial.pt. OR random:.tw. OR tu.xs. #18 animals/ NOT human/ #19 17 NOT 18 #20 13 AND 16 AND 19

EMBASE #1 Heparin/ #2 heparin.tw #3 Low Molecular Weight Heparin/ #4 (LMWH OR low molecular weight heparin OR nadroparin OR fraxiparin OR enoxaparin OR clexane OR lovenox OR dalteparin OR fragmin OR ardeparin OR normiflo OR tinzaparin OR logi- parin OR innohep OR certoparin OR sandoparin OR reviparin OR clivarin OR danaproid OR orgaran).tw #51OR2OR3OR4 #6 Coumarin derivative/ #7 Warfarin/ #8 (warfarin OR coumadin OR acenocumarol OR phenprocumon OR 4-hydroxicoumarins OR oral anticoagulant OR vitamin K an- tagonist OR VKA).tw #96OR7OR8 #10 fondaparinux/ #11 (fondaparinux OR Arixtra).tw #12 ximelagatran/ #13 (ximelagatran OR Exanta).tw #14 (Pradaxa OR Dabigatran OR rivaroxaban OR Xarelto OR apix- aban).tw. #155OR9OR10OR11OR12OR13OR14 #16 Neoplasm/ #17 (malignan$ OR neoplasm$ OR cancer OR carcinoma$ OR adenocarcinoma OR tumour OR tumor).tw #18 16 OR 17 #19 Random:.tw. OR clinical trial:.mp. OR exp health care quality #20 animals/ NOT human/ #21 19 NOT 20 #22 15 AND 18 AND 21

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 54 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. (Continued)

ISI (International Scientific Information) the Web of Science #1 heparin OR low molecular weight heparin OR LMWH OR low- molecular-weight-heparin OR nadroparin OR fraxiparin OR enoxa- parin OR clexane OR lovenox OR dalteparin OR fragmin OR arde- parin OR normiflo OR tinzaparin OR logiparin OR innohep OR certoparin OR sandoparin OR reviparin OR clivarin OR danaproid OR orgaran #2 Coumarins OR Warfarin OR coumadin OR acenocumarol OR phenprocumon OR 4-hydroxicoumarins OR oral anticoagulant OR vitamin K antagonist OR VKA #3 fondaparinux OR Arixtra #4 ximelagatran OR Exanta # 5 Pradaxa OR Dabigatran OR rivaroxaban OR Xarelto OR apix- aban #61OR2OR3OR4OR5 #7 malignan$ OR neoplasm$ OR cancer OR carcinoma$ OR ade- nocarcinoma OR tumour OR tumor #8 random$ OR placebo$ OR versus OR vs OR double blind OR double-blind OR compar$ OR controlled #96AND7AND8

CENTRAL (The Cochrane Library, latest issue) #1 heparin OR low molecular weight heparin OR LMWH OR low- molecular-weight-heparin OR nadroparin OR fraxiparin OR enoxa- parin OR clexane OR lovenox OR dalteparin OR fragmin OR arde- parin OR normiflo OR tinzaparin OR logiparin OR innohep OR certoparin OR sandoparin OR reviparin OR clivarin OR danaproid OR orgaran #2 Coumarins OR Warfarin OR coumadin OR acenocumarol OR phenprocumon OR 4-hydroxicoumarins OR oral anticoagulant OR vitamin K antagonist OR VKA #3 fondaparinux OR Arixtra #4 ximelagatran OR Exanta #5 Pradaxa or Dabigatran or rivaroxaban or Xarelto or apixaban #61OR2OR3OR4OR5 #7 malignan$ OR neoplasm$ OR cancer OR carcinoma$ OR ade- nocarcinoma OR tumour OR tumor #8 6AND7

WHAT’S NEW Last assessed as up-to-date: 5 December 2010.

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 55 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Date Event Description

6 December 2010 New search has been performed Text revised

7 February 2010 New citation required but conclusions have not Updated search, February 2010. changed

HISTORY Review first published: Issue 3, 2007

Date Event Description

15 May 2007 New citation required and conclusions have changed Substantive amendment We updated the classification of heterogeneity. We consid- ered the following classification of heterogeneity based on the value of I2: 0% to 30% = low; 30% to 60% = moderate and worthy of investigation; 60% to 90% = severe and wor- thy of understanding; 90% to 100% = allowing aggregation only with major caution (Julian Higgins, personal commu- nication). We rephrased the abstract conclusion as follows: “This review suggests a survival benefit of heparin in cancer patients in general, and in patients with small cell lung cancer in particular.” We also added the following to the 4th paragraph of the Discussion (“Interpretation of the findings of this review is limited by the moderate heterogeneity…”): “The interpretation of findings is also limited by not includ- ing data from the seven trials published as abstracts only.”

CONTRIBUTIONSOFAUTHORS EAA: protocol development, search for trials, screening, data extraction, data analysis, manuscript drafting, review co-ordination. SKG: search for trials, screening, full text retrieval, data extraction, data analysis, manuscript drafting. MB: screening. VY: full text retrieval, data extraction. FFvD: data extraction, data analysis. SK: data extraction, data analysis. SM: data extraction, data analysis. HOD: statistical analysis, methodological advice, data extraction, manuscript drafting. AB: statistical analysis. HJS: protocol development, search for trials, data analysis, methodological expertise.

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 56 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. DECLARATIONSOFINTEREST HJS: no personal payments from for-profit sponsors related to the subject matter in the past three years. EAA and HJS are executive committee members of the ACCP Antithrombotic Therapy Guidelines.

SOURCES OF SUPPORT

Internal sources • State University of New York at Buffalo, Department of Medicine, USA. • Italian National Cancer Institute Regina Elena, Rome, Italy. • Academic Medical Center, Department of Vascular Medicine, Netherlands.

External sources • Research Grants, Not specified. H Schünemann: no personal payments from for-profit sponsors. Research grants and honoraria were received into research accounts or received by a research group that he belongs to from AstraZeneca, Amgen, Chiesi Foundation, Lily, Pfizer, Roche and UnitedBioSource for development or consulting regarding quality of life instruments for chronic respiratory diseases and as lecture fees related to the methodology of evidence-based practice guideline development and research methodology. Institutions or organizations that he is affiliated with likely receive funding from for-profit sponsors that are supporting infrastructure and research that may serve his work.

DIFFERENCESBETWEENPROTOCOLANDREVIEW We intended to use continuous data for quality of life outcomes but no eligible study assessed this outcome.

INDEX TERMS

Medical Subject Headings (MeSH) Anticoagulants [∗administration & dosage; adverse effects]; Carcinoma, Small Cell [mortality]; Hemorrhage [chemically induced]; Hep- arin [∗administration & dosage; adverse effects]; Heparin, Low-Molecular-Weight [administration & dosage]; Neoplasms [∗mortality]; Randomized Controlled Trials as Topic; Survival Analysis; Time Factors; Venous Thromboembolism [∗prevention & control]; Warfarin [administration & dosage]

MeSH check words Humans

Parenteral anticoagulation in patients with cancer who have no therapeutic or prophylactic indication for anticoagulation (Review) 57 Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.