Adverse Events

On commence par charger la base TrialMaster comme on fait d’habitude. Ici j’utilise la fonction edc_example_ae() pour ne pas dépendre de données réelles privées.

library(EDCimport)
library(flextable)
# tm = read_trialmaster("path/to/file.zip")
tm = edc_example_ae()
load_list(tm)

head(ae)
#> # A tibble: 6 × 6
#>   subjid aesoc                                          aegr  n_ae sae   crfname
#>    <int> <chr>                                         <int> <int> <lgl> <chr>  
#> 1      1 Nervous system disorders                          3     2 FALSE ae     
#> 2      1 Injury, poisoning and procedural complicatio…     1     2 FALSE ae     
#> 3      2 Hepatobiliary disorders                           1     2 TRUE  ae     
#> 4      2 Injury, poisoning and procedural complicatio…     2     2 FALSE ae     
#> 5      3 Blood and lymphatic system disorders              4     3 FALSE ae     
#> 6      3 Injury, poisoning and procedural complicatio…     5     3 FALSE ae
head(enrolres)
#> # A tibble: 6 × 3
#>   subjid arm   crfname 
#>    <int> <chr> <chr>   
#> 1      1 Trt   enrolres
#> 2      2 Trt   enrolres
#> 3      3 Trt   enrolres
#> 4      4 Trt   enrolres
#> 5      5 Ctl   enrolres
#> 6      6 Ctl   enrolres

Macro AE_grades

La macro AE_grades est traduite en deux fonctions : ae_table_grade_max() et ae_table_grade_n(). Ces fonctions retournent des objets de classe crosstable. Ils ont une méthode as_flextable qui les transforme en table HTML de classe flextable.

EDCimport comprend aussi la fonction ae_plot_grade_max()

Voir la description du package crosstable pour plus d’informations: documentation.
Voir la description du package flextable pour la liste des modificateurs (comme add_footer_lines()) : documentation.

Tableaux

• AE_grades1
• AE_grades2
• AE_grades3
• AE_grades4

• Table des grades maximum par patient

  ae_table_grade_max(df_ae=ae, df_enrol=enrolres, arm=NULL) %>% 
    as_flextable() %>% 
    add_footer_lines("Percentages are given as the proportion of patients presenting at most one 
                     AE of given grade")

  label	variable	value
  Max grade	Grade 1	4 (8%)
  	Grade 2	10 (20%)
  	Grade 3	18 (36%)
  	Grade 4	9 (18%)
  	Grade 5	6 (12%)
  	No AE	3 (6%)
  	Total	50 (100%)
  Percentages are given as the proportion of patients presenting at most one AE of given grade

Table des grades maximum par patient stratifié sur le bras

On pourrait retrouver exactement la sortie SAS en mettant total=FALSE.

ae_table_grade_max(df_ae=ae, df_enrol=enrolres, arm="arm") %>% 
  as_flextable() %>% 
  add_footer_lines("Percentages are given as the proportion of patients presenting at most one AE of given grade")

label	variable	arm_		Total
		Ctl	Trt
Max grade	Grade 1	3 (11%)	1 (4%)	4 (8%)
	Grade 2	5 (19%)	5 (22%)	10 (20%)
	Grade 3	10 (37%)	8 (35%)	18 (36%)
	Grade 4	8 (30%)	1 (4%)	9 (18%)
	Grade 5	1 (4%)	5 (22%)	6 (12%)
	No AE	0 (0%)	3 (13%)	3 (6%)
	Total	27 (54%)	23 (46%)	50 (100%)
Percentages are given as the proportion of patients presenting at most one AE of given grade

Table de tous les grades pour chaque patient

ae_table_grade_n(df_ae=ae, df_enrol=enrolres, arm=NULL) %>% 
  as_flextable() %>% 
  add_footer_lines("Percentages are given as the proportion of patients presenting at least one AE of given grade")

label	variable	Treatment arm
		All patients
grade	Grade 1	28 (56%)
	Grade 2	25 (50%)
	Grade 3	27 (54%)
	Grade 4	11 (22%)
	Grade 5	6 (12%)
Percentages are given as the proportion of patients presenting at least one AE of given grade

On pourrait retrouver exactement la sortie SAS en mettant total=FALSE.

ae_table_grade_n(df_ae=ae, df_enrol=enrolres, arm="arm") %>% 
  as_flextable() %>% 
  add_footer_lines("Percentages are given as the proportion of patients presenting at least one AE of given grade")

label	variable	arm
		Ctl	Trt
grade	Grade 1	15 (56%)	13 (57%)
	Grade 2	17 (63%)	8 (35%)
	Grade 3	14 (52%)	13 (57%)
	Grade 4	8 (30%)	3 (13%)
	Grade 5	1 (4%)	5 (22%)
Percentages are given as the proportion of patients presenting at least one AE of given grade

Table des grades maximum par patient, filtrée sur les SAE

La fonction ne prend pas en charge les SAE directement, il suffit de filtrer la table AE en amont.

ae %>% 
  dplyr::filter(sae==TRUE) %>% 
  ae_table_grade_max(df_enrol=enrolres, arm=NULL) %>% 
  as_flextable() %>% 
  add_footer_lines("Percentages are given as the proportion of patients presenting at most one SAE of given grade")

label	variable	value
Max grade	Grade 1	4 (8%)
	Grade 2	3 (6%)
	Grade 3	2 (4%)
	Grade 4	2 (4%)
	Grade 5	1 (2%)
	No AE	38 (76%)
	Total	50 (100%)
Percentages are given as the proportion of patients presenting at most one SAE of given grade

Sorties Plots

• AE max
• AE N

• ae_plot_grade_max(df_ae=ae, df_enrol=enrolres, 
                    type = c("stack", "dodge", "fill")) & 
    ggplot2::labs(fill="Arm")
  #> Warning: Returning more (or less) than 1 row per `summarise()` group was deprecated in
  #> dplyr 1.1.0.
  #> ℹ Please use `reframe()` instead.
  #> ℹ When switching from `summarise()` to `reframe()`, remember that `reframe()`
  #>   always returns an ungrouped data frame and adjust accordingly.
  #> ℹ The deprecated feature was likely used in the EDCimport package.
  #>   Please report the issue at <https://github.com/DanChaltiel/EDCimport/issues>.

  

ae_plot_grade_n(df_ae=ae, df_enrol=enrolres)

Macro AE_SOC

La macro AE_SOC est traduite en la fonctions : ae_table_soc(). Cette fonction retourne des objets de classe ae_table_soc. Ils ont également une méthode as_flextable qui les transforme en table HTML de classe flextable, mais différente de celle de crosstable.

Voir la description du package flextable pour la liste des modificateurs (comme add_footer_lines()) : documentation.

Pour les sorties sur officer, comme ces tables sont très larges, pensez bien à basculer en format paysage en utilisant officer::body_end_section_continuous(), puis officer::body_end_section_landscape() pour revenir au format portrait.

Tableaux

• AE_SOC1
• AE_SOC2
• AE_SOC3
• AE_SOC4/5
• AE_SOC6

• Table des grades par soc et termes

  On peut ajouter total=FALSE pour retirer la colonne “Tot”.

  ae_table_soc(df_ae=ae, df_enrol=enrolres, term=NULL, arm=NULL, sort_by_ae=FALSE) %>% 
    as_flextable() %>% 
    add_footer_lines("In the header, N represents the number of patients.") %>% 
    add_footer_lines("Percentages are given as the proportion of patients presenting at least one AE of given grade")

  	All patients (N=50)
  CTCAE SOC	G1	G2	G3	G4	G5	Tot
  Blood and lymphatic system disorders	3 (6%)	1 (2%)		1 (2%)		5 (10%)
  Cardiac disorders	1 (2%)	2 (4%)	1 (2%)	1 (2%)		5 (10%)
  Ear and labyrinth disorders	2 (4%)	1 (2%)	1 (2%)	2 (4%)		6 (12%)
  Endocrine disorders		3 (6%)		1 (2%)		4 (8%)
  Eye disorders	3 (6%)	1 (2%)	2 (4%)	1 (2%)		7 (14%)
  Gastrointestinal disorders		3 (6%)	1 (2%)		1 (2%)	5 (10%)
  General disorders and administration site conditions	3 (6%)	1 (2%)	1 (2%)			5 (10%)
  Hepatobiliary disorders	2 (4%)	3 (6%)	1 (2%)	1 (2%)		7 (14%)
  Immune system disorders	1 (2%)		1 (2%)	1 (2%)	1 (2%)	4 (8%)
  Infections and infestations	2 (4%)	2 (4%)	2 (4%)		1 (2%)	7 (14%)
  Injury, poisoning and procedural complications	4 (8%)	1 (2%)	2 (4%)	1 (2%)	1 (2%)	9 (18%)
  Investigations	2 (4%)	2 (4%)		1 (2%)		5 (10%)
  Metabolism and nutrition disorders		1 (2%)				1 (2%)
  Musculoskeletal and connective tissue disorders	3 (6%)	2 (4%)				5 (10%)
  Neoplasms benign, malignant and unspecified (incl cysts and polyps)	4 (8%)	1 (2%)	2 (4%)		1 (2%)	8 (16%)
  Nervous system disorders	1 (2%)	1 (2%)	6 (12%)			8 (16%)
  Psychiatric disorders	2 (4%)	1 (2%)	2 (4%)			5 (10%)
  Renal and urinary disorders		1 (2%)	1 (2%)	1 (2%)		3 (6%)
  Reproductive system and breast disorders	2 (4%)	4 (8%)				6 (12%)
  Respiratory, thoracic and mediastinal disorders	2 (4%)		3 (6%)	1 (2%)		6 (12%)
  Skin and subcutaneous tissue disorders		2 (4%)	3 (6%)	2 (4%)		7 (14%)
  Surgical and medical procedures	3 (6%)	1 (2%)	1 (2%)			5 (10%)
  Vascular disorders		2 (4%)	2 (4%)		1 (2%)	5 (10%)
  In the header, N represents the number of patients.
  Percentages are given as the proportion of patients presenting at least one AE of given grade

Table des grades par soc et termes

##TODO exemple avec des TERMS

able des grades par termes uniquement, filtrée sur les SAE

##TODO ajouter vrais SAE

Table des grades stratifiée sur le bras (sans colonne total)

On peut ajouter total=FALSE pour retirer la colonne “Tot” et obtenir la sortie AE_SOC5.

ae_table_soc(df_ae=ae, df_enrol=enrolres, term=NULL, arm="arm", sort_by_ae=FALSE) %>% 
  as_flextable() %>% 
  add_footer_lines("In the header, N represents the number of patients.") %>% 
  add_footer_lines("Percentages are given as the proportion of patients presenting at least one AE of given grade")

	Ctl (N=27)						Trt (N=23)
CTCAE SOC	G1	G2	G3	G4	G5	Tot	G1	G2	G3	G4	G5	Tot
Blood and lymphatic system disorders	2 (7%)	1 (4%)				3 (11%)	1 (4%)			1 (4%)		2 (9%)
Cardiac disorders	1 (4%)	1 (4%)	1 (4%)	1 (4%)		4 (15%)		1 (4%)				1 (4%)
Ear and labyrinth disorders	1 (4%)					1 (4%)	1 (4%)	1 (4%)	1 (4%)	2 (9%)		5 (22%)
Endocrine disorders		3 (11%)		1 (4%)		4 (15%)
Eye disorders	2 (7%)		2 (7%)	1 (4%)		5 (19%)	1 (4%)	1 (4%)				2 (9%)
Gastrointestinal disorders		2 (7%)	1 (4%)		1 (4%)	4 (15%)		1 (4%)				1 (4%)
General disorders and administration site conditions	3 (11%)		1 (4%)			4 (15%)		1 (4%)				1 (4%)
Hepatobiliary disorders		2 (7%)		1 (4%)		3 (11%)	2 (9%)	1 (4%)	1 (4%)			4 (17%)
Immune system disorders	1 (4%)		1 (4%)	1 (4%)		3 (11%)					1 (4%)	1 (4%)
Infections and infestations	2 (7%)	2 (7%)	1 (4%)			5 (19%)			1 (4%)		1 (4%)	2 (9%)
Injury, poisoning and procedural complications	1 (4%)		1 (4%)	1 (4%)		3 (11%)	3 (13%)	1 (4%)	1 (4%)		1 (4%)	6 (26%)
Investigations	1 (4%)	1 (4%)		1 (4%)		3 (11%)	1 (4%)	1 (4%)				2 (9%)
Metabolism and nutrition disorders		1 (4%)				1 (4%)
Musculoskeletal and connective tissue disorders	1 (4%)	1 (4%)				2 (7%)	2 (9%)	1 (4%)				3 (13%)
Neoplasms benign, malignant and unspecified (incl cysts and polyps)	2 (7%)	1 (4%)				3 (11%)	2 (9%)		2 (9%)		1 (4%)	5 (22%)
Nervous system disorders	1 (4%)	1 (4%)	3 (11%)			5 (19%)			3 (13%)			3 (13%)
Psychiatric disorders	1 (4%)	1 (4%)	1 (4%)			3 (11%)	1 (4%)		1 (4%)			2 (9%)
Renal and urinary disorders		1 (4%)	1 (4%)	1 (4%)		3 (11%)
Reproductive system and breast disorders	2 (7%)	4 (15%)				6 (22%)
Respiratory, thoracic and mediastinal disorders	2 (7%)		1 (4%)	1 (4%)		4 (15%)			2 (9%)			2 (9%)
Skin and subcutaneous tissue disorders			1 (4%)	1 (4%)		2 (7%)		2 (9%)	2 (9%)	1 (4%)		5 (22%)
Surgical and medical procedures	1 (4%)	1 (4%)				2 (7%)	2 (9%)		1 (4%)			3 (13%)
Vascular disorders		1 (4%)	2 (7%)			3 (11%)		1 (4%)			1 (4%)	2 (9%)
In the header, N represents the number of patients.
Percentages are given as the proportion of patients presenting at least one AE of given grade

Table des soc et termes tous grades confondus stratifiée sur le bras

##TODO

Sortie Plots

• Butterfly plot

• Ici, on considère comme sévères les AE de grade >= 3 :

  ae %>% 
    dplyr::mutate(severe = aegr>=3) %>% 
    ae_plot_soc(df_enrol=enrolres)