- Email: [email protected]
Reporting infections in clinical trials of patients with haematological malignancies
Clinical Microbiology and Infection xxx (xxxx) xxx
Contents lists available at ScienceDirect
Clinical Microbiology and Infection journal homepage: www.clinicalmicrobiologyandinfection.com
Systematic review
Reporting infections in clinical trials of patients with haematological malignancies N. Tau 1, 2, y, L. Shargian-Alon 2, 3, S. Reich 4, M. Paul 5, 6, A. Gafter-Gvili 2, 3, 4, D. Shepshelovich 2, 4, *, D. Yahav 2, 7, * 1
Department of Diagnostic Imaging, Chaim Sheba Medical Centre, Ramat Gan, Israel Sackler School of Medicine, Tel Aviv University, Ramat Aviv, Israel Institute of Haematology, Davidoff Cancer Centre, Rabin Medical Centre, Petah-Tikva, Israel 4 Medicine A, Rabin Medical Centre, Beilinson Hospital, Petah Tikva, Israel 5 Infectious Diseases Unit, Rambam Health Care Campus, Haifa, Israel 6 Faculty of Medicine, Technion, Israel Institute of Technology, Haifa, Israel 7 Infectious Diseases Unit, Rabin Medical Centre, Beilinson Hospital, Petah Tikva, Israel 2 3
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 March 2019 Received in revised form 24 April 2019 Accepted 28 April 2019 Available online xxx
Background: Infections are common among patients treated for haematological malignancies and are associated with significant morbidity and mortality. The completeness of reporting infectious complications in randomized controlled trials (RCTs) assessing treatments for haematological malignancies is unknown. Objectives: We aimed to evaluate the completeness of reporting infectious complications in RCTs assessing treatments for haematological malignancies. Data source: A systematic literature search was performed in PubMed database. Study eligibility criteria and participants: All primary published phase II/III RCTs between September 2016 and September 2018 evaluating treatments for haematological malignancies in adult patients were included. Intervention: Reporting infectious complications. Methods: A systematic review was conducted to evaluate the completeness of reporting. Study characteristics and data concerning reporting of infectious complications were collected by two independent reviewers. Quality of reporting was assessed using a modification of the CONSORT extension checklist for harms, including 15 items. Results: One-hundred and seven RCTs were included. Most trials (97; 91%) provided some report on infections. Approximately half reported on each of pneumonia, sepsis and neutropenic fever; 12 trials (11%) reported on fungal infections. Only nine trials (8%) listed infections by type of pathogen (i.e. bacterial, fungal or viral) and 48 (45%) by source/type of infection (i.e. pneumonia, urinary tract infection, etc.). Most trials did not address infections in their title, abstract, introduction or discussion. Median number of items of the CONSORT modification reported was 7 points, (interquartile range (IQR) 6e9) for all included trials, with lower median for 34 acute leukaemia trials (median 6, IQR 5e8). Conclusions: Most trials evaluating treatment for haematological malignancies provide some data relating to infectious complications. The reports are mostly incomplete and rarely provided in a structured presentation. N. Tau, Clin Microbiol Infect 2019;▪:1 © 2019 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Editor: C. Pulcini Keywords: CONSORT CONSORT harms Haematological malignancies Randomized controlled trials
Introduction y Corresponding author. N. Tau, Department of Diagnostic Imaging, The Chaim Sheba Medical Centre, Tel Hashomer, 5265601, Ramat Gan, Israel. E-mail address: [email protected] (N. Tau). * D. Shepshelovich and D. Yahav contributed equally.
Infections are common in individuals treated for haematological malignancies and are associated with major clinical implications [1]. The toxicity burden of infectious complications among these
https://doi.org/10.1016/j.cmi.2019.04.029 1198-743X/© 2019 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
2
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
patients might surpass that associated with the direct complications from the treatment of the haematological malignancy [2]. Some studies reported infections to be the direct cause of death in up to 60% of patients with haematological neoplasms [3,4]. For some anti-cancer drugs, data regarding treatment-related infections have only been reported post-marketing. Many drugs for haematological malignancies are approved for marketing based on single-arm trials with a small sample size, making it difficult to attribute uncommon infectious complications to the studied drug [5]. However, data regarding clinical syndromes and causative agents of infectious complications are sometimes also lacking from randomized controlled trials (RCTs) [6]. In the case of Ibrutinib, a considerable rate (4%) of invasive aspergillosis has been reported in post-marketing data, but was unnoticed in preceding clinical trials. This may be due to the relatively small sample size of the trials or lack of a specific pathogen being reported in many trials; some trials only reported ‘pneumonia’ as an infectious complication without specifying the causative pathogen, while others only reported ‘infections’ [7]. Reporting the causative organisms and infection sources associated with specific treatment regimens may influence the appropriate prophylactic antimicrobial therapy and the empirical treatment of clinical infections in patients with haematological malignancy. We aimed to evaluate the completeness of reporting infectious complications in RCTs assessing treatment for haematological malignancies.
malignancies were also excluded. For the detailed search strategy see Appendix 1. Data extraction and reporting completeness assessment Full text and supplementary materials of included trials were inspected and data were abstracted by two of four independent reviewers (DY, DS, LS, SR). Discrepancies were resolved through a discussion with a third reviewer (NT). Extracted data included: publication journal and 2017 impact factor, year of publication, source of funding (industry or nonindustry), type of haematological malignancy, name of studied drugs, sample size and duration of follow up. Infection reporting data were also collected, and divided into two categoriesdquantitative evaluation of reported infectious complications and qualitative evaluation of reporting, which was assessed using the CONSORT group harms extension checklist modified to include infectious complications (Table 1) [10]. In addition to the CONSORT harms extension checklist, we also counted the most relevant criteria concerning infection reporting summing to a maximum of 15 items [11,12]. The criteria used and their description are detailed in Table 2. In addition, we searched for trials' protocols in clinical trial databases and examined discrepancies in reporting infection between registry and final publication. Statistical analysis
Methods A systematic review of peer-reviewed literature was performed. For this review, the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines were followed [8]. PubMed was searched for phase II/III trials published between 1 September 2016 and 1 September 2018 reporting treatment of haematological malignancies. Three reviewers (NT, DY or DS) screened the titles obtained from the search and identified RCTs comparing various treatment regimens versus another regimen, placebo or supportive care. Inclusion criteria were limited to English-language publications; trials including adult patients (>18 years); and recent publications (between September 2016 and September 2018). This recent time period was chosen because reporting of clinical trials is constantly evolving and improving [9], and therefore we aimed to assess recent reports to create a more homogeneous cohort and prevent potential bias. Treatments for prevention of graft-versus-host disease, tumour lysis, mucositis or infectious complications were excluded. Phase I trials as well as trials addressing benign haematological disorders or solid
We collected and summarized data into tables, including characteristics of trials, type of infections reported and quality of reporting. The latter was measured using a score designed by the authors for the current review, in the absence of a validated tool to address specifically infections. The score included 15 items, each receiving one point, with a score of 15 corresponding to optimal reporting (for score items see Table 2). Frequencies are expressed as numbers and percentages (%). Descriptive statistical analysis was performed using SPSS version 25.0 (Chicago, IL, USA). Sensitivity analyses were conducted according to type of malignancy (acute leukaemia/myelodysplastic syndrome versus others); and according to impact factor of the publishing journal (high impact factor, defined as >10, versus lower impact factor). Results Study selection flow chart is presented in Fig. 1. Overall, 107 RCTs reporting results for 32 843 patients were included. Ninety-eight (92%) were multicentre studies and 68 (64%) were industry funded. Thirty-seven (35%) trials were phase II
Table 1 Reporting of infectious complications Variable name
Definition
Number (%) trials reporting (n ¼ 107)
Infections reported
Were infections in general (i.e. any infection) reported in the publication? Were pneumonia episodes reported? Were sepsis or septic shock episodes reported? Were neutropenic fever episodes reported? Were fungal infections reported? Were infectious complications classified to summarize number according to pathogen type? (number of bacterial, fungal or viral infections) Did the study results favour the investigated drug/regimen? (As was concluded by study's authors)
97 (91%)
Pneumonia reported Sepsis/septic shock reported Neutropenic fever reported Fungal infections reported Infections classified by pathogen type Positive study results
58 (54%) 52 (49%) 59 (55%) 12 (11%) 9 (8%)
71/104 (68%)
Any mention of the above infections was considered as ‘yes’, even without full reporting of number of episodes.
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
and 70 (65%) were phase III trials. Median impact factor of journals of publication was 10.7, range 0e79.3; 71 (66%) trials were published in journals with impact factor >10. Most trials (n ¼ 73, 68%) were concluded by the authors as having positive results. The most common malignancies were acute myeloid leukaemia (n ¼ 25, 23%) and non-Hodgkin lymphoma (n ¼ 25, 23%). Others were chronic lymphocytic lymphoma (n ¼ 17, 16%); multiple myeloma (n ¼ 16, 15%); myeloproliferative neoplasms (n ¼ 8, 7%); myelodysplastic syndrome (n ¼ 5, 5%), acute lymphocytic leukaemia (n ¼ 4, 4%); Hodgkin disease (n ¼ 4, 4%) and chronic myeloid leukaemia (n ¼ 3, 3%). The regimens of treatment are depicted in the supplementary material (Table S1).
3
General reporting rates of infections Most trials (97/107, 91%) provided some report on infections (Table 1). Approximately half of the trials reported on each of the following; pneumonia, sepsis/septic shock and neutropenic fever. These reports included any reference to the infection, and did not necessarily report all cases of such infection. Only nine trials (8%) provided details on the type of pathogen involved in the infection (i.e. bacterial, fungal or viral). Only 12 trials (11%) provided any report on fungal infections, with none of the other trials declaring that no fungal infections were documented. Fifty-seven trials (53%) reported infectious-complications-related data in the main manuscript, 5 (5%) in the supplementary data section and 37 (35%) both in the main manuscript and supplementary data section.
Table 2 CONSORT harm modification for reporting infectious adverse events Variable name
Definition
Number (%) trials reporting (n ¼ 107)
Infections reported in the title or abstract*
Were infections mentioned in title or abstract of the publication? Were infections mentioned in the introduction of the publication?
19 (18%)
Introduction* Definition of infections a. Definition of infection*
b. Definition of severity*
c. All or selected samples *
d. Treatment emergent adverse events Collection of harm-related data: a. Mode of data collection*
b. Methods to attribute harm to intervention* c. Stopping rules Plans for presenting and analysin information on harms: a. Description of plans b. Grouping of infections*
c. Handling of recurrent infections* Death due to infection* Denominators for analysed of harms: a. Number analysed* b. Definition of analysis set*
Present the absolute risk of each infectious adverse event a. Severity and grading of infections* b. Number of patients/events c. Results of infections presented separately for each treatment arm* Discussion*
a. Is definition for the diagnosis of various infection provided (for example CDC criteria for diagnosis of pneumonia, urinary tract infections) b. Was a standardized and validated instrument used for grading of severity of AEs? (CTCAE and WHO were considered validated instrument) c. Did authors clarify whether the reported AEs encompass all the recorded AEs or a selected sample with description of the criteria for the selected sample (i.e. reporting only AEs with a predefined severity or frequencY d. Were TEAEs addressed in the publication?
5 (5%)
a. 0 (0%)
b. 81 (76%)
c. 93 (87%)
d. 39 (36%)
a. Did authors describe how data were collected? (visit, questionnaires, interviews etc.) b. Was a description of how AEs were attributed to trial drugs provided? c. Were stopping rules for the trial pre-defined?
a. 49 (46%)
a. Description of methods for presenting and/or analysing AEs in general is provided b. Were reports of infections grouped by type/source of infection (e.g. pneumonia, skin and soft-tissue infection) c. Were plans how to handle recurrent events provided? Were number of deaths due to infection reported per study arm?
a. 13 (12%)
b. 17 (16%) c. 33 (31%)
b. 48 (45%)
c. 7 (7%) 69 (64%)
a. Were the denominators for any AEs provided per study arm? b. Did authors provide definition for analysis set? (describe whether AEs reported by intention-to-treat or per-protocol or other population)
a. 97 (91%)
a. Were severe infections reported separately? b. Were both number of patients with infection and number of infectious episodes reported? c. Were the denominators for infections provided per study arm?
a. 92 (86%) b. 7 (7%)
Were infections addressed in the discussion of the publication?
30 (28%)
b. 85 (79%)
c. 92 (86%)
AEs, adverse events; CTCAE, Common Terminology Criteria for Adverse Events; TEAE, treatment emergent adverse events. *Items marked with an asterisk were those counted as important items for reporting, summing in maximum 15 items.
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
4
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
Fig. 1. Study selection flow chart.
CONSORT harm modification for reporting infections Quality of infectious complications reporting assessed by the CONSORT harms checklist is presented in Table 2. Most trials did not address infections in their title, abstract, introduction or discussion, and only reported infectious complications in the results section. Criteria for the diagnosis of type/source of infection (e.g. pneumonia, urinary tract infection, upper respiratory tract infection) were not explicitly defined in any of the trials. However, a validated instrument for grading severity of infection) such as Common Terminology Criteria for Adverse Events and WHO) was used in most trials [13,14] and most reported separately on severe infections (as defined by the prescribed instrument) separately. Less than half of the trials provided data on methods of infectious-complications-related information collection. In addition, only 7% of trials reported recurrent events and subsequently, only these trials presented data separately for patients and infectious episodes. Presentation of infections by grouping into type/source of infection was available for 45% of trials. Data on infection-related mortality was reported on a perarm basis in 64% of trials (for detailed data see Table 1). The duration of follow up for infectious complications was specified in 21 studies (20%), with a median of 3 months (range
1e72 months). Counting items reported of 15 itemsdmedian reporting was 7, interquartile range (IQR) 6e9 (Table 2). Individual study scores according to type of haematological disease and according to study phase are presented in the Supplementary material (Tables S2 and S3, respectively). Sensitivity analysis including 34 trials addressing acute leukaemia/myelodysplastic syndrome versus other diseases demonstrated lower reporting rates of infections in these trials (median 6 items reported, IQR 5e8 compared with median of 8, IQR 7e9 in 73 trials addressing other malignancies, p 0.005). In the trials addressing acute leukaemia/myelodysplastic syndrome, only 13 trials (38%) reported occurrence of any pneumonia in these patients; 12 trials (35%) reported sepsis/septic shock; and 6 trials (18%) reported fungal infections. Categorizing the infections by type of pathogen was performed in 4 trials (12%); and grouping by type/source of infection in 12 trials (35%). Infections were mentioned in the discussion section of only a quarter of acute leukaemia trials. Sensitivity analysis by publishing journal impact factor did not demonstrate a significant difference in reporting rate of 71 publications in journals with impact factor >10 versus lower impact factor (median 8, IQR 6e9 versus median 7, IQR 6e8.75, p 0.294). Studies published in higher impact factor journals were
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
significantly more likely to be multi-centred and industryfunded. These trials reported better on methods of harmrelated data collection, listed infections by type of pathogen in 11% of trials (versus 3% with lower impact factor), and notably, provided report on fungal infections in 12/71 trials (17%) compared with no report on fungal infections in 36 trials published in journal with impact factor <10. Discrepancies in reporting infection between registry and publication Of the 107 included trials, 100 were registered in a clinical trial registry. Of the registered trials, 95 did not define infections as an end point (85/95 (89%) reported infection-related outcomes in the final publication), whereas all five trial registries that defined infections among the outcomes planned reported infections in the published manuscript. Seven trials were not registered. Discussion In 107 RCTs (32 843 participants) addressing various treatment regimens for haematological malignancies, we found heterogeneous quality of reporting of infectious events. Using a score of 1 to 15 designed for evaluating quality of reporting based on CONSORT harmdmedian score among these 107 RCTs was 7, with IQR 6e9. Though in general most trials provided some data relating to infections in the results section or supplementary material, these were mostly incomplete and were rarely provided in a structured presentation. Reporting of infections was not properly designed in most trials, reflected by the fact that standardized definitions for infections were not used in any of the trials; recurrent infectious episodes were not addressed; and no methodology for reporting different pathogen groups was reported. As a result, most trials did not provide data for infections classified by either pathogen type or source of infection. Infections are common in patients treated for haematological malignancies, and constitute a leading cause of morbidity and mortality [3,4]. It has been suggested that patients enrolled in clinical trials may even have higher risk for infections, perhaps reflecting the more intensive therapeutic regimens offered [15]. Interestingly, quality of specifically reporting infections in trials addressing acute leukaemia was surprisingly inferior to that for other malignancies, although patients with this malignancy are at higher risk for such complications. Reporting harms in oncology trials has been criticized for under-reporting and downplaying the seriousness of adverse events. The risk in under-reporting may be over-estimation of the riskebenefit ratio of a new cancer drug [16,17]. A recent review by Phillips et al. looked at the approach to collection, reporting and analysis of adverse events in RCTs [18]. The review found significant inconsistencies and underuse of adverse events reporting and analysis in clinical trials. In the RCTs reviewed in our study, adverse events in general were defined and reported, usually the more common or severe ones. Infections, however, were not specifically addressed among them. Failure to completely report and discuss infectious complications of new treatments for haematological malignancies might bias the perception of the efficacy and safety balance of the assessed therapies by under-estimating the related toxicities. Underreporting of infections in published trials may lead to
5
observing the occurrence of life-threatening infections only at post-marketing phase [19]. Complete reporting of sources of infection and causative organisms associated with specific treatment regimens may inform physicians regarding appropriate antimicrobial prophylaxis and empiric treatments. This can be achieved by aggregate reporting of infections according to source and isolated pathogens. A suggested structure for reporting is detailed in Appendix 2. Unfortunately, a recent report found aggregate reporting of any kind was used by only 29% of phase 3 oncology trials [20]. In conclusion, infectious complications associated with treatments for haematological malignancies are incompletely reported. Health-care professionals, journal editors and reviewers should demand complete and transparent reporting of infectious complications in trials assessing new treatments for haematological malignancies. A suggested structure for reporting is detailed in Appendix 2. Infections should be reported in the population of all patients randomized and receiving the intervention, without a minimal incidence threshold, to enable early detection of rare infections. Accepted definitions for diagnosing these infections should be used [21].
Transparency declaration All authors have declared that there are no conflicts of interest in association with this article.
Funding No external funding was received.
Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.cmi.2019.04.029.
Appendix 1. Search strategy The search term: (‘hematological neoplasms’ [MeSH] or lymphoproliferative disorders or lymphoma or leukemia or non?Hodgkin lymphoma or Hodgkin lymphoma or plasma cell dyscrasia or multiple myeloma or myeloproliferative neoplasms or polycythemia vera or chronic myeloid leukemia or myelodysplastic syndrome or acute leukemia or acute myeloid leukemia or acute lymphatic leukemia or chronic lymphocytic leukemia or mantle cell lymphoma or mycosis fungoides) was combined with the Cochrane filter for randomized controlled trials ((randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized [tiab] OR placebo [tiab] OR clinical trials as topic [mesh: noexp] OR randomly [tiab] OR trial [ti]) NOT (animals [mh] NOT humans [mh])) PubMed filters were used to limit for age (19þ years) and publication dates (2 years up to 15 September 2018).
Appendix 2. Suggested table for collection of infectious adverse events data Reporting of all positive adverse events should be mandatory. Main types of infections and pathogens should be reported, suggested as follows:
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
6
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
Source of infection
Number of patients
Number of episodes
Number with grade 3 severity according to CTCAE
Lower respiratory tract infections: Bacterial Pseudomonas aeruginosa Nocardia spp. Fungal Aspergillus spp. Zygomycetes spp. Pneumocystis jirovecii Cryptococcus neoformans Viral Cytomegalovirus Mycobacterial Mycobacterium tuberculosis Unspecified Bloodstream infections: Bacterial Type of bacteria (elaborate) Fungal Candida spp. Other (elaborate) Sinusitis Bacterial Fungal Aspergillus spp. Zygomycetes spp. Urinary tract infections Skin and soft-tissue infections Bacterial Upper respiratory tract infections Viral Influenza Respiratory syncytial virus Others Intra-abdominal infections Enterocolitis Bacterial Viral Clostridium difficile Cytomegalovirus Typhlitis Peritonitis Cholecystitis Perianal infection Other Herpes simplex virus infection Mucocutaneous Disseminated Herpes zoster virus Disseminated herpes zoster Cytomegalovirus infection Viraemia Cytomegalovirus syndrome Central nervous system infection Bacterial meningitis Listeria monocytogenes Viral encephalitis Herpes simplex virus 1 Human herpesvirus 6 West-Nile virus JC virus (progressive multifocal leukoencephalopathy) Fungal Aspergillus spp. Cryptococcal meningitis Parasitic infections Toxoplasmosis (specify site) Strongyloides stercoralisddisseminated Other infections
References [1] Zaorsky NG, Churilla TM, Egleston BL, Fisher SG, Ridge JA, Horwitz EM, et al. Causes of death among cancer patients. Ann Oncol 2017;28:400e7. [2] Lee SM, Hershman DL, Miao J, Zhong X, Unger JM, Cheung YKK. Estimating global treatment toxicity burden from adverse-event data. Cancer 2018;124: 858e64.
[3] Chang HY, Rodriguez V, Narboni G, Bodey GP, Luna MA, Freireich EJ. Causes of death in adults with acute leukemia. Medicine (Balt.) 1976;55:259e68. [4] Nosari A, Barberis M, Landonio G, Magnani P, Majno M, Oreste P, et al. Infections in haematologic neoplasms: autopsy findings. Haematologica 1991;76:135e40. [5] Shepshelovich D, Tibau A, Goldvaser H, Ocana A, Seruga B, Amir E. Postmarketing safety-related modifications of drugs approved by the US Food and
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
[6]
[7]
[8]
[9]
[10]
[11]
Drug Administration between 1999 and 2014 without randomized controlled trials. Mayo Clin Proc 2019;94:74e83. Fernandez-Ruiz M, Meije Y, Manuel O, Akan H, Carratala J, Aguado JM, et al. ESCMID study group for infections in compromised hosts (ESGICH) consensus document on the safety of targeted and biological therapies: an infectious diseases perspective (introduction). Clin Microbiol Infect 2018;24:S2e9. Dreyling M, Jurczak W, Jerkeman M, Silva RS, Rusconi C, Trneny M, et al. Ibrutinib versus temsirolimus in patients with relapsed or refractory mantlecell lymphoma: an international, randomised, open-label, phase 3 study. Lancet 2016;387:770e8. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol 2009;62:1006e12. Shepshelovich D, Yelin D, Gafter-Gvili A, Goldman S, Avni T, Yahav D. Comparison of reporting phase III randomized controlled trials of antibiotic treatment for common bacterial infections in ClinicalTrials.gov and matched publications. Clin Microbiol Infect 2018;24. 1211 e9ee14. Ioannidis JP, Evans SJ, Gotzsche PC, O'Neill RT, Altman DG, Schulz K, et al. Better reporting of harms in randomized trials: an extension of the CONSORT statement. Ann Intern Med 2004;141:781e8. Peron J, Maillet D, Gan HK, Chen EX, You B. Adherence to CONSORT adverse event reporting guidelines in randomized clinical trials evaluating systemic cancer therapy: a systematic review. J Clin Oncol 2013;31:3957e63.
7
[12] Haidich AB, Birtsou C, Dardavessis T, Tirodimos I, Arvanitidou M. The quality of safety reporting in trials is still suboptimal: survey of major general medical journals. J Clin Epidemiol 2011;64:124e35. [13] National Cancer Institute. Common Terminology criteria for adverse events v5.0 (CTCAE). Bethesda, MD: NCI; 2017. [14] World Health Organization. WHO toxicity grades. Geneva: WHO. [15] Dix D, Aplenc R, Bowes L, Cellot S, Ethier MC, Feusner J, et al. Impact of registration on clinical trials on infection risk in pediatric acute myeloid leukemia. Int J Cancer 2016;138:1785e91. [16] Gyawali B, Shimokata T, Honda K, Ando Y. Reporting harms more transparently in trials of cancer drugs. BMJ 2018;363:k4383. [17] Seruga B, Templeton AJ, Badillo FE, Ocana A, Amir E, Tannock IF. Underreporting of harm in clinical trials. Lancet Oncol 2016;17:e209e19. [18] Phillips R, Hazell L, Sauzet O, Cornelius V. Analysis and reporting of adverse events in randomised controlled trials: a review. BMJ Open 2019;9:e024537. [19] Carson KR, Newsome SD, Kim EJ, Wagner-Johnston ND, von Geldern G, Moskowitz CH, et al. Progressive multifocal leukoencephalopathy associated with brentuximab vedotin therapy: a report of 5 cases from the Southern Network on Adverse Reactions (SONAR) project. Cancer 2014;120:2464e71. [20] Maillet D, Gan HK, Blay JY, You B, Peron J. Aggregated adverse-events outcomes in oncology phase III reports: a systematic review. Eur J Cancer 2016;52:26e32. [21] CDC/NHSN. Highlighted NHSN january 2013 patient safety component (PSC) manual updates 2013 21-Apr-2019. Atlanta, GA: CDC.
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
Contents lists available at ScienceDirect
Clinical Microbiology and Infection journal homepage: www.clinicalmicrobiologyandinfection.com
Systematic review
Reporting infections in clinical trials of patients with haematological malignancies N. Tau 1, 2, y, L. Shargian-Alon 2, 3, S. Reich 4, M. Paul 5, 6, A. Gafter-Gvili 2, 3, 4, D. Shepshelovich 2, 4, *, D. Yahav 2, 7, * 1
Department of Diagnostic Imaging, Chaim Sheba Medical Centre, Ramat Gan, Israel Sackler School of Medicine, Tel Aviv University, Ramat Aviv, Israel Institute of Haematology, Davidoff Cancer Centre, Rabin Medical Centre, Petah-Tikva, Israel 4 Medicine A, Rabin Medical Centre, Beilinson Hospital, Petah Tikva, Israel 5 Infectious Diseases Unit, Rambam Health Care Campus, Haifa, Israel 6 Faculty of Medicine, Technion, Israel Institute of Technology, Haifa, Israel 7 Infectious Diseases Unit, Rabin Medical Centre, Beilinson Hospital, Petah Tikva, Israel 2 3
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 March 2019 Received in revised form 24 April 2019 Accepted 28 April 2019 Available online xxx
Background: Infections are common among patients treated for haematological malignancies and are associated with significant morbidity and mortality. The completeness of reporting infectious complications in randomized controlled trials (RCTs) assessing treatments for haematological malignancies is unknown. Objectives: We aimed to evaluate the completeness of reporting infectious complications in RCTs assessing treatments for haematological malignancies. Data source: A systematic literature search was performed in PubMed database. Study eligibility criteria and participants: All primary published phase II/III RCTs between September 2016 and September 2018 evaluating treatments for haematological malignancies in adult patients were included. Intervention: Reporting infectious complications. Methods: A systematic review was conducted to evaluate the completeness of reporting. Study characteristics and data concerning reporting of infectious complications were collected by two independent reviewers. Quality of reporting was assessed using a modification of the CONSORT extension checklist for harms, including 15 items. Results: One-hundred and seven RCTs were included. Most trials (97; 91%) provided some report on infections. Approximately half reported on each of pneumonia, sepsis and neutropenic fever; 12 trials (11%) reported on fungal infections. Only nine trials (8%) listed infections by type of pathogen (i.e. bacterial, fungal or viral) and 48 (45%) by source/type of infection (i.e. pneumonia, urinary tract infection, etc.). Most trials did not address infections in their title, abstract, introduction or discussion. Median number of items of the CONSORT modification reported was 7 points, (interquartile range (IQR) 6e9) for all included trials, with lower median for 34 acute leukaemia trials (median 6, IQR 5e8). Conclusions: Most trials evaluating treatment for haematological malignancies provide some data relating to infectious complications. The reports are mostly incomplete and rarely provided in a structured presentation. N. Tau, Clin Microbiol Infect 2019;▪:1 © 2019 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Editor: C. Pulcini Keywords: CONSORT CONSORT harms Haematological malignancies Randomized controlled trials
Introduction y Corresponding author. N. Tau, Department of Diagnostic Imaging, The Chaim Sheba Medical Centre, Tel Hashomer, 5265601, Ramat Gan, Israel. E-mail address: [email protected] (N. Tau). * D. Shepshelovich and D. Yahav contributed equally.
Infections are common in individuals treated for haematological malignancies and are associated with major clinical implications [1]. The toxicity burden of infectious complications among these
https://doi.org/10.1016/j.cmi.2019.04.029 1198-743X/© 2019 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
2
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
patients might surpass that associated with the direct complications from the treatment of the haematological malignancy [2]. Some studies reported infections to be the direct cause of death in up to 60% of patients with haematological neoplasms [3,4]. For some anti-cancer drugs, data regarding treatment-related infections have only been reported post-marketing. Many drugs for haematological malignancies are approved for marketing based on single-arm trials with a small sample size, making it difficult to attribute uncommon infectious complications to the studied drug [5]. However, data regarding clinical syndromes and causative agents of infectious complications are sometimes also lacking from randomized controlled trials (RCTs) [6]. In the case of Ibrutinib, a considerable rate (4%) of invasive aspergillosis has been reported in post-marketing data, but was unnoticed in preceding clinical trials. This may be due to the relatively small sample size of the trials or lack of a specific pathogen being reported in many trials; some trials only reported ‘pneumonia’ as an infectious complication without specifying the causative pathogen, while others only reported ‘infections’ [7]. Reporting the causative organisms and infection sources associated with specific treatment regimens may influence the appropriate prophylactic antimicrobial therapy and the empirical treatment of clinical infections in patients with haematological malignancy. We aimed to evaluate the completeness of reporting infectious complications in RCTs assessing treatment for haematological malignancies.
malignancies were also excluded. For the detailed search strategy see Appendix 1. Data extraction and reporting completeness assessment Full text and supplementary materials of included trials were inspected and data were abstracted by two of four independent reviewers (DY, DS, LS, SR). Discrepancies were resolved through a discussion with a third reviewer (NT). Extracted data included: publication journal and 2017 impact factor, year of publication, source of funding (industry or nonindustry), type of haematological malignancy, name of studied drugs, sample size and duration of follow up. Infection reporting data were also collected, and divided into two categoriesdquantitative evaluation of reported infectious complications and qualitative evaluation of reporting, which was assessed using the CONSORT group harms extension checklist modified to include infectious complications (Table 1) [10]. In addition to the CONSORT harms extension checklist, we also counted the most relevant criteria concerning infection reporting summing to a maximum of 15 items [11,12]. The criteria used and their description are detailed in Table 2. In addition, we searched for trials' protocols in clinical trial databases and examined discrepancies in reporting infection between registry and final publication. Statistical analysis
Methods A systematic review of peer-reviewed literature was performed. For this review, the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines were followed [8]. PubMed was searched for phase II/III trials published between 1 September 2016 and 1 September 2018 reporting treatment of haematological malignancies. Three reviewers (NT, DY or DS) screened the titles obtained from the search and identified RCTs comparing various treatment regimens versus another regimen, placebo or supportive care. Inclusion criteria were limited to English-language publications; trials including adult patients (>18 years); and recent publications (between September 2016 and September 2018). This recent time period was chosen because reporting of clinical trials is constantly evolving and improving [9], and therefore we aimed to assess recent reports to create a more homogeneous cohort and prevent potential bias. Treatments for prevention of graft-versus-host disease, tumour lysis, mucositis or infectious complications were excluded. Phase I trials as well as trials addressing benign haematological disorders or solid
We collected and summarized data into tables, including characteristics of trials, type of infections reported and quality of reporting. The latter was measured using a score designed by the authors for the current review, in the absence of a validated tool to address specifically infections. The score included 15 items, each receiving one point, with a score of 15 corresponding to optimal reporting (for score items see Table 2). Frequencies are expressed as numbers and percentages (%). Descriptive statistical analysis was performed using SPSS version 25.0 (Chicago, IL, USA). Sensitivity analyses were conducted according to type of malignancy (acute leukaemia/myelodysplastic syndrome versus others); and according to impact factor of the publishing journal (high impact factor, defined as >10, versus lower impact factor). Results Study selection flow chart is presented in Fig. 1. Overall, 107 RCTs reporting results for 32 843 patients were included. Ninety-eight (92%) were multicentre studies and 68 (64%) were industry funded. Thirty-seven (35%) trials were phase II
Table 1 Reporting of infectious complications Variable name
Definition
Number (%) trials reporting (n ¼ 107)
Infections reported
Were infections in general (i.e. any infection) reported in the publication? Were pneumonia episodes reported? Were sepsis or septic shock episodes reported? Were neutropenic fever episodes reported? Were fungal infections reported? Were infectious complications classified to summarize number according to pathogen type? (number of bacterial, fungal or viral infections) Did the study results favour the investigated drug/regimen? (As was concluded by study's authors)
97 (91%)
Pneumonia reported Sepsis/septic shock reported Neutropenic fever reported Fungal infections reported Infections classified by pathogen type Positive study results
58 (54%) 52 (49%) 59 (55%) 12 (11%) 9 (8%)
71/104 (68%)
Any mention of the above infections was considered as ‘yes’, even without full reporting of number of episodes.
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
and 70 (65%) were phase III trials. Median impact factor of journals of publication was 10.7, range 0e79.3; 71 (66%) trials were published in journals with impact factor >10. Most trials (n ¼ 73, 68%) were concluded by the authors as having positive results. The most common malignancies were acute myeloid leukaemia (n ¼ 25, 23%) and non-Hodgkin lymphoma (n ¼ 25, 23%). Others were chronic lymphocytic lymphoma (n ¼ 17, 16%); multiple myeloma (n ¼ 16, 15%); myeloproliferative neoplasms (n ¼ 8, 7%); myelodysplastic syndrome (n ¼ 5, 5%), acute lymphocytic leukaemia (n ¼ 4, 4%); Hodgkin disease (n ¼ 4, 4%) and chronic myeloid leukaemia (n ¼ 3, 3%). The regimens of treatment are depicted in the supplementary material (Table S1).
3
General reporting rates of infections Most trials (97/107, 91%) provided some report on infections (Table 1). Approximately half of the trials reported on each of the following; pneumonia, sepsis/septic shock and neutropenic fever. These reports included any reference to the infection, and did not necessarily report all cases of such infection. Only nine trials (8%) provided details on the type of pathogen involved in the infection (i.e. bacterial, fungal or viral). Only 12 trials (11%) provided any report on fungal infections, with none of the other trials declaring that no fungal infections were documented. Fifty-seven trials (53%) reported infectious-complications-related data in the main manuscript, 5 (5%) in the supplementary data section and 37 (35%) both in the main manuscript and supplementary data section.
Table 2 CONSORT harm modification for reporting infectious adverse events Variable name
Definition
Number (%) trials reporting (n ¼ 107)
Infections reported in the title or abstract*
Were infections mentioned in title or abstract of the publication? Were infections mentioned in the introduction of the publication?
19 (18%)
Introduction* Definition of infections a. Definition of infection*
b. Definition of severity*
c. All or selected samples *
d. Treatment emergent adverse events Collection of harm-related data: a. Mode of data collection*
b. Methods to attribute harm to intervention* c. Stopping rules Plans for presenting and analysin information on harms: a. Description of plans b. Grouping of infections*
c. Handling of recurrent infections* Death due to infection* Denominators for analysed of harms: a. Number analysed* b. Definition of analysis set*
Present the absolute risk of each infectious adverse event a. Severity and grading of infections* b. Number of patients/events c. Results of infections presented separately for each treatment arm* Discussion*
a. Is definition for the diagnosis of various infection provided (for example CDC criteria for diagnosis of pneumonia, urinary tract infections) b. Was a standardized and validated instrument used for grading of severity of AEs? (CTCAE and WHO were considered validated instrument) c. Did authors clarify whether the reported AEs encompass all the recorded AEs or a selected sample with description of the criteria for the selected sample (i.e. reporting only AEs with a predefined severity or frequencY d. Were TEAEs addressed in the publication?
5 (5%)
a. 0 (0%)
b. 81 (76%)
c. 93 (87%)
d. 39 (36%)
a. Did authors describe how data were collected? (visit, questionnaires, interviews etc.) b. Was a description of how AEs were attributed to trial drugs provided? c. Were stopping rules for the trial pre-defined?
a. 49 (46%)
a. Description of methods for presenting and/or analysing AEs in general is provided b. Were reports of infections grouped by type/source of infection (e.g. pneumonia, skin and soft-tissue infection) c. Were plans how to handle recurrent events provided? Were number of deaths due to infection reported per study arm?
a. 13 (12%)
b. 17 (16%) c. 33 (31%)
b. 48 (45%)
c. 7 (7%) 69 (64%)
a. Were the denominators for any AEs provided per study arm? b. Did authors provide definition for analysis set? (describe whether AEs reported by intention-to-treat or per-protocol or other population)
a. 97 (91%)
a. Were severe infections reported separately? b. Were both number of patients with infection and number of infectious episodes reported? c. Were the denominators for infections provided per study arm?
a. 92 (86%) b. 7 (7%)
Were infections addressed in the discussion of the publication?
30 (28%)
b. 85 (79%)
c. 92 (86%)
AEs, adverse events; CTCAE, Common Terminology Criteria for Adverse Events; TEAE, treatment emergent adverse events. *Items marked with an asterisk were those counted as important items for reporting, summing in maximum 15 items.
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
4
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
Fig. 1. Study selection flow chart.
CONSORT harm modification for reporting infections Quality of infectious complications reporting assessed by the CONSORT harms checklist is presented in Table 2. Most trials did not address infections in their title, abstract, introduction or discussion, and only reported infectious complications in the results section. Criteria for the diagnosis of type/source of infection (e.g. pneumonia, urinary tract infection, upper respiratory tract infection) were not explicitly defined in any of the trials. However, a validated instrument for grading severity of infection) such as Common Terminology Criteria for Adverse Events and WHO) was used in most trials [13,14] and most reported separately on severe infections (as defined by the prescribed instrument) separately. Less than half of the trials provided data on methods of infectious-complications-related information collection. In addition, only 7% of trials reported recurrent events and subsequently, only these trials presented data separately for patients and infectious episodes. Presentation of infections by grouping into type/source of infection was available for 45% of trials. Data on infection-related mortality was reported on a perarm basis in 64% of trials (for detailed data see Table 1). The duration of follow up for infectious complications was specified in 21 studies (20%), with a median of 3 months (range
1e72 months). Counting items reported of 15 itemsdmedian reporting was 7, interquartile range (IQR) 6e9 (Table 2). Individual study scores according to type of haematological disease and according to study phase are presented in the Supplementary material (Tables S2 and S3, respectively). Sensitivity analysis including 34 trials addressing acute leukaemia/myelodysplastic syndrome versus other diseases demonstrated lower reporting rates of infections in these trials (median 6 items reported, IQR 5e8 compared with median of 8, IQR 7e9 in 73 trials addressing other malignancies, p 0.005). In the trials addressing acute leukaemia/myelodysplastic syndrome, only 13 trials (38%) reported occurrence of any pneumonia in these patients; 12 trials (35%) reported sepsis/septic shock; and 6 trials (18%) reported fungal infections. Categorizing the infections by type of pathogen was performed in 4 trials (12%); and grouping by type/source of infection in 12 trials (35%). Infections were mentioned in the discussion section of only a quarter of acute leukaemia trials. Sensitivity analysis by publishing journal impact factor did not demonstrate a significant difference in reporting rate of 71 publications in journals with impact factor >10 versus lower impact factor (median 8, IQR 6e9 versus median 7, IQR 6e8.75, p 0.294). Studies published in higher impact factor journals were
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
significantly more likely to be multi-centred and industryfunded. These trials reported better on methods of harmrelated data collection, listed infections by type of pathogen in 11% of trials (versus 3% with lower impact factor), and notably, provided report on fungal infections in 12/71 trials (17%) compared with no report on fungal infections in 36 trials published in journal with impact factor <10. Discrepancies in reporting infection between registry and publication Of the 107 included trials, 100 were registered in a clinical trial registry. Of the registered trials, 95 did not define infections as an end point (85/95 (89%) reported infection-related outcomes in the final publication), whereas all five trial registries that defined infections among the outcomes planned reported infections in the published manuscript. Seven trials were not registered. Discussion In 107 RCTs (32 843 participants) addressing various treatment regimens for haematological malignancies, we found heterogeneous quality of reporting of infectious events. Using a score of 1 to 15 designed for evaluating quality of reporting based on CONSORT harmdmedian score among these 107 RCTs was 7, with IQR 6e9. Though in general most trials provided some data relating to infections in the results section or supplementary material, these were mostly incomplete and were rarely provided in a structured presentation. Reporting of infections was not properly designed in most trials, reflected by the fact that standardized definitions for infections were not used in any of the trials; recurrent infectious episodes were not addressed; and no methodology for reporting different pathogen groups was reported. As a result, most trials did not provide data for infections classified by either pathogen type or source of infection. Infections are common in patients treated for haematological malignancies, and constitute a leading cause of morbidity and mortality [3,4]. It has been suggested that patients enrolled in clinical trials may even have higher risk for infections, perhaps reflecting the more intensive therapeutic regimens offered [15]. Interestingly, quality of specifically reporting infections in trials addressing acute leukaemia was surprisingly inferior to that for other malignancies, although patients with this malignancy are at higher risk for such complications. Reporting harms in oncology trials has been criticized for under-reporting and downplaying the seriousness of adverse events. The risk in under-reporting may be over-estimation of the riskebenefit ratio of a new cancer drug [16,17]. A recent review by Phillips et al. looked at the approach to collection, reporting and analysis of adverse events in RCTs [18]. The review found significant inconsistencies and underuse of adverse events reporting and analysis in clinical trials. In the RCTs reviewed in our study, adverse events in general were defined and reported, usually the more common or severe ones. Infections, however, were not specifically addressed among them. Failure to completely report and discuss infectious complications of new treatments for haematological malignancies might bias the perception of the efficacy and safety balance of the assessed therapies by under-estimating the related toxicities. Underreporting of infections in published trials may lead to
5
observing the occurrence of life-threatening infections only at post-marketing phase [19]. Complete reporting of sources of infection and causative organisms associated with specific treatment regimens may inform physicians regarding appropriate antimicrobial prophylaxis and empiric treatments. This can be achieved by aggregate reporting of infections according to source and isolated pathogens. A suggested structure for reporting is detailed in Appendix 2. Unfortunately, a recent report found aggregate reporting of any kind was used by only 29% of phase 3 oncology trials [20]. In conclusion, infectious complications associated with treatments for haematological malignancies are incompletely reported. Health-care professionals, journal editors and reviewers should demand complete and transparent reporting of infectious complications in trials assessing new treatments for haematological malignancies. A suggested structure for reporting is detailed in Appendix 2. Infections should be reported in the population of all patients randomized and receiving the intervention, without a minimal incidence threshold, to enable early detection of rare infections. Accepted definitions for diagnosing these infections should be used [21].
Transparency declaration All authors have declared that there are no conflicts of interest in association with this article.
Funding No external funding was received.
Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.cmi.2019.04.029.
Appendix 1. Search strategy The search term: (‘hematological neoplasms’ [MeSH] or lymphoproliferative disorders or lymphoma or leukemia or non?Hodgkin lymphoma or Hodgkin lymphoma or plasma cell dyscrasia or multiple myeloma or myeloproliferative neoplasms or polycythemia vera or chronic myeloid leukemia or myelodysplastic syndrome or acute leukemia or acute myeloid leukemia or acute lymphatic leukemia or chronic lymphocytic leukemia or mantle cell lymphoma or mycosis fungoides) was combined with the Cochrane filter for randomized controlled trials ((randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized [tiab] OR placebo [tiab] OR clinical trials as topic [mesh: noexp] OR randomly [tiab] OR trial [ti]) NOT (animals [mh] NOT humans [mh])) PubMed filters were used to limit for age (19þ years) and publication dates (2 years up to 15 September 2018).
Appendix 2. Suggested table for collection of infectious adverse events data Reporting of all positive adverse events should be mandatory. Main types of infections and pathogens should be reported, suggested as follows:
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
6
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
Source of infection
Number of patients
Number of episodes
Number with grade 3 severity according to CTCAE
Lower respiratory tract infections: Bacterial Pseudomonas aeruginosa Nocardia spp. Fungal Aspergillus spp. Zygomycetes spp. Pneumocystis jirovecii Cryptococcus neoformans Viral Cytomegalovirus Mycobacterial Mycobacterium tuberculosis Unspecified Bloodstream infections: Bacterial Type of bacteria (elaborate) Fungal Candida spp. Other (elaborate) Sinusitis Bacterial Fungal Aspergillus spp. Zygomycetes spp. Urinary tract infections Skin and soft-tissue infections Bacterial Upper respiratory tract infections Viral Influenza Respiratory syncytial virus Others Intra-abdominal infections Enterocolitis Bacterial Viral Clostridium difficile Cytomegalovirus Typhlitis Peritonitis Cholecystitis Perianal infection Other Herpes simplex virus infection Mucocutaneous Disseminated Herpes zoster virus Disseminated herpes zoster Cytomegalovirus infection Viraemia Cytomegalovirus syndrome Central nervous system infection Bacterial meningitis Listeria monocytogenes Viral encephalitis Herpes simplex virus 1 Human herpesvirus 6 West-Nile virus JC virus (progressive multifocal leukoencephalopathy) Fungal Aspergillus spp. Cryptococcal meningitis Parasitic infections Toxoplasmosis (specify site) Strongyloides stercoralisddisseminated Other infections
References [1] Zaorsky NG, Churilla TM, Egleston BL, Fisher SG, Ridge JA, Horwitz EM, et al. Causes of death among cancer patients. Ann Oncol 2017;28:400e7. [2] Lee SM, Hershman DL, Miao J, Zhong X, Unger JM, Cheung YKK. Estimating global treatment toxicity burden from adverse-event data. Cancer 2018;124: 858e64.
[3] Chang HY, Rodriguez V, Narboni G, Bodey GP, Luna MA, Freireich EJ. Causes of death in adults with acute leukemia. Medicine (Balt.) 1976;55:259e68. [4] Nosari A, Barberis M, Landonio G, Magnani P, Majno M, Oreste P, et al. Infections in haematologic neoplasms: autopsy findings. Haematologica 1991;76:135e40. [5] Shepshelovich D, Tibau A, Goldvaser H, Ocana A, Seruga B, Amir E. Postmarketing safety-related modifications of drugs approved by the US Food and
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029
N. Tau et al. / Clinical Microbiology and Infection xxx (xxxx) xxx
[6]
[7]
[8]
[9]
[10]
[11]
Drug Administration between 1999 and 2014 without randomized controlled trials. Mayo Clin Proc 2019;94:74e83. Fernandez-Ruiz M, Meije Y, Manuel O, Akan H, Carratala J, Aguado JM, et al. ESCMID study group for infections in compromised hosts (ESGICH) consensus document on the safety of targeted and biological therapies: an infectious diseases perspective (introduction). Clin Microbiol Infect 2018;24:S2e9. Dreyling M, Jurczak W, Jerkeman M, Silva RS, Rusconi C, Trneny M, et al. Ibrutinib versus temsirolimus in patients with relapsed or refractory mantlecell lymphoma: an international, randomised, open-label, phase 3 study. Lancet 2016;387:770e8. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol 2009;62:1006e12. Shepshelovich D, Yelin D, Gafter-Gvili A, Goldman S, Avni T, Yahav D. Comparison of reporting phase III randomized controlled trials of antibiotic treatment for common bacterial infections in ClinicalTrials.gov and matched publications. Clin Microbiol Infect 2018;24. 1211 e9ee14. Ioannidis JP, Evans SJ, Gotzsche PC, O'Neill RT, Altman DG, Schulz K, et al. Better reporting of harms in randomized trials: an extension of the CONSORT statement. Ann Intern Med 2004;141:781e8. Peron J, Maillet D, Gan HK, Chen EX, You B. Adherence to CONSORT adverse event reporting guidelines in randomized clinical trials evaluating systemic cancer therapy: a systematic review. J Clin Oncol 2013;31:3957e63.
7
[12] Haidich AB, Birtsou C, Dardavessis T, Tirodimos I, Arvanitidou M. The quality of safety reporting in trials is still suboptimal: survey of major general medical journals. J Clin Epidemiol 2011;64:124e35. [13] National Cancer Institute. Common Terminology criteria for adverse events v5.0 (CTCAE). Bethesda, MD: NCI; 2017. [14] World Health Organization. WHO toxicity grades. Geneva: WHO. [15] Dix D, Aplenc R, Bowes L, Cellot S, Ethier MC, Feusner J, et al. Impact of registration on clinical trials on infection risk in pediatric acute myeloid leukemia. Int J Cancer 2016;138:1785e91. [16] Gyawali B, Shimokata T, Honda K, Ando Y. Reporting harms more transparently in trials of cancer drugs. BMJ 2018;363:k4383. [17] Seruga B, Templeton AJ, Badillo FE, Ocana A, Amir E, Tannock IF. Underreporting of harm in clinical trials. Lancet Oncol 2016;17:e209e19. [18] Phillips R, Hazell L, Sauzet O, Cornelius V. Analysis and reporting of adverse events in randomised controlled trials: a review. BMJ Open 2019;9:e024537. [19] Carson KR, Newsome SD, Kim EJ, Wagner-Johnston ND, von Geldern G, Moskowitz CH, et al. Progressive multifocal leukoencephalopathy associated with brentuximab vedotin therapy: a report of 5 cases from the Southern Network on Adverse Reactions (SONAR) project. Cancer 2014;120:2464e71. [20] Maillet D, Gan HK, Blay JY, You B, Peron J. Aggregated adverse-events outcomes in oncology phase III reports: a systematic review. Eur J Cancer 2016;52:26e32. [21] CDC/NHSN. Highlighted NHSN january 2013 patient safety component (PSC) manual updates 2013 21-Apr-2019. Atlanta, GA: CDC.
Please cite this article as: Tau N et al., Reporting infections in clinical trials of patients with haematological malignancies, Clinical Microbiology and Infection, https://doi.org/10.1016/j.cmi.2019.04.029