Skip to contents

A4 and LEARN Screening and Baseline

ATRI Biostatistics

Source: vignettes/A4-LEARN-Baseline-Characteristics.Rmd
A4-LEARN-Baseline-Characteristics.Rmd

The Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease (A4) study was a Phase 3, double-blind, placebo-controlled study of solanezumab in subjects with preclinical Alzheimer’s Disease (Sperling et al. 2023). The Longitudinal Evaluation of Amyloid Risk and Neurodegeneration (LEARN) study enrolled clinically unimpaired individuals who screen-failed for the A4 Study on the basis of not showing elevated amyloid on screening amyloid PET imaging and followed them longitudinally with all of the same assessments as A4.

The code below demonstrates how to reproduce Table 1 from Sperling et al. (2020) using data in the A4LEARN package and R Core Team (2024).

Load required R packages

Prepare data

Rescreens

Participants who re-screened may appear in the data twice with different BIDs each time. The A4LEARN::SUBJINFO derived dataset indicates which participants are re-screens, and how the re-screen BIDs are mapped to each other. The code below also accounts for this to set up the removal of duplicate appearances.

rescreens <- A4LEARN::SUBJINFO %>%
  filter(!is.na(PREVBID)) %>%
  rename(BID1 = PREVBID, BID2 = BID) %>%
  select(BID1, BID2)

Gather Amyloid PET data

The Amyloid PET quantitative data (A4LEARN::imaging_SUVR_amyloid) is in long format with one row per region. We use tidyr::pivot_wider to transform to wide format with one column per region.

pet <- A4LEARN::imaging_SUVR_amyloid %>% 
  filter(brain_region != '' & VISCODE == 2) %>%
  pivot_wider(id_cols='BID', names_from=brain_region, values_from=suvr_cer) %>%
  left_join(A4LEARN::imaging_PET_VA, by='BID') %>%
  left_join(rescreens, by=c('BID' = 'BID1')) %>% 
  mutate( # update PET BIDs to second BID if necessary
    BID = case_when(
      !is.na(BID2) ~ BID2,
      TRUE ~ BID)) %>%
  arrange(BID, BID2) %>%
  filter(!duplicated(BID, fromLast = TRUE)) %>%
  select(-BID2)

Baseline PACC

pacc_bl <- A4LEARN::PACC %>% 
  filter(VISCODE == 6) %>%
  select(BID, PACC.raw, MMSCORE, LDELTOTAL, DIGITTOTAL, FCTOTAL96)

Prepare data table

dd <- A4LEARN::SUBJINFO %>% select(BID, APOEGN) %>%
  left_join(A4LEARN::ptdemog, by='BID') %>%
  left_join(pet, by=c('SUBSTUDY','BID')) %>%
  left_join(pacc_bl, by=c('BID')) %>%
  # remove first appearance of rescreens:
  filter(!BID %in% rescreens$BID1) %>% 
  rename(`A4 amyloid eligibility` = overall_score) %>%
  rename(
    `Age at screening (yrs)` = PTAGE,
    `APOE genotype` = APOEGN, 
    `Education (yrs)` = PTEDUCAT, 
    `Amyloid PET SUVr^b^` = Composite_Summary,
    Sex = PTGENDER,
    Ethnicity = PTETHNIC,
    `Marital status` = PTMARRY,
    `Participant retired` = PTNOTRT,
    PACC = PACC.raw, 
    MMSE = MMSCORE, 
    `Logical memory delay` = LDELTOTAL, 
    `Digit symbol` = DIGITTOTAL, 
    `FCSRT (2xFree + Cued)` = FCTOTAL96,
    `A4 amyloid eligibility^b^` = `A4 amyloid eligibility`) %>% 
  # rename values and fields for footnote setup in summary table
  mutate(SUBSTUDY = factor(SUBSTUDY, levels = c('A4', 'LEARN'),
    labels = c('A4^a^' , 'LEARN')))

Characteristics by screening amyloid PET status 

tableby(`A4 amyloid eligibility^b^` ~ .,
  data = dd %>%
    filter(!is.na(`Amyloid PET SUVr^b^`)) %>%
    select(`A4 amyloid eligibility^b^`, `Age at screening (yrs)`, 
      `Education (yrs)`, `APOE genotype`, `Amyloid PET SUVr^b^`, 
      Sex, Ethnicity, `Marital status`, `Participant retired`, PACC, MMSE, 
      `Logical memory delay`, `Digit symbol`, `FCSRT (2xFree + Cued)`), 
  digits = 2) %>%
  summary(title = "Characteristics of all who underwent screening amyloid PET with comparison of Not Elevated (Aβ−) and Elevated Amyloid (Aβ+) groups")
Characteristics of all who underwent screening amyloid PET with comparison of Not Elevated (Aβ−) and Elevated Amyloid (Aβ+) groups
Negative (N=3163) Positive (N=1323) Total (N=4486) p value
Age at screening (yrs) < 0.001
   Mean (SD) 70.95 (4.53) 72.10 (4.89) 71.29 (4.67)
   Range 64.63 - 85.93 65.00 - 85.74 64.63 - 85.93
Education (yrs) 0.535
   N-Miss 4 2 6
   Mean (SD) 16.60 (2.85) 16.54 (2.81) 16.58 (2.84)
   Range 8.00 - 32.00 7.00 - 30.00 7.00 - 32.00
APOE genotype < 0.001
   N-Miss 30 2 32
   E2/E2 23 (0.7%) 2 (0.2%) 25 (0.6%)
   E2/E3 380 (12.1%) 69 (5.2%) 449 (10.1%)
   E2/E4 74 (2.4%) 42 (3.2%) 116 (2.6%)
   E3/E3 1938 (61.9%) 482 (36.5%) 2420 (54.3%)
   E3/E4 684 (21.8%) 618 (46.8%) 1302 (29.2%)
   E4/E4 34 (1.1%) 108 (8.2%) 142 (3.2%)
Amyloid PET SUVrb < 0.001
   Mean (SD) 0.99 (0.07) 1.33 (0.18) 1.09 (0.19)
   Range 0.70 - 1.59 0.97 - 2.09 0.70 - 2.09
Sex 0.623
   Male 1278 (40.4%) 545 (41.2%) 1823 (40.6%)
   Female 1885 (59.6%) 778 (58.8%) 2663 (59.4%)
Ethnicity 0.193
   Hispanic or Latino 103 (3.3%) 39 (2.9%) 142 (3.2%)
   Not Hispanic or Latino 3040 (96.1%) 1269 (95.9%) 4309 (96.1%)
   Unknown or Not reported 20 (0.6%) 15 (1.1%) 35 (0.8%)
Marital status 0.635
   Married 2223 (70.3%) 943 (71.3%) 3166 (70.6%)
   Widowed 304 (9.6%) 122 (9.2%) 426 (9.5%)
   Divorced 438 (13.8%) 190 (14.4%) 628 (14.0%)
   Never married 135 (4.3%) 48 (3.6%) 183 (4.1%)
   Unknown/Other 63 (2.0%) 20 (1.5%) 83 (1.9%)
Participant retired 0.942
   N-Miss 1 0 1
   Yes 2396 (75.8%) 1005 (76.0%) 3401 (75.8%)
   No 724 (22.9%) 299 (22.6%) 1023 (22.8%)
   Not Applicable 42 (1.3%) 19 (1.4%) 61 (1.4%)
PACC < 0.001
   N-Miss 2625 151 2776
   Mean (SD) 0.79 (2.35) -0.00 (2.68) 0.25 (2.60)
   Range -8.70 - 6.64 -12.52 - 7.75 -12.52 - 7.75
MMSE < 0.001
   N-Miss 2625 151 2776
   Mean (SD) 29.03 (1.17) 28.78 (1.28) 28.86 (1.25)
   Range 23.00 - 30.00 22.00 - 30.00 22.00 - 30.00
Logical memory delay < 0.001
   N-Miss 2625 151 2776
   Mean (SD) 13.54 (3.35) 12.62 (3.68) 12.91 (3.60)
   Range 3.00 - 24.00 0.00 - 23.00 0.00 - 24.00
Digit symbol 0.010
   N-Miss 2625 151 2776
   Mean (SD) 49.97 (9.89) 48.64 (10.02) 49.06 (9.99)
   Range 0.00 - 79.00 15.00 - 86.00 0.00 - 86.00
FCSRT (2xFree + Cued) < 0.001
   N-Miss 2625 151 2776
   Mean (SD) 78.67 (5.83) 77.33 (6.29) 77.76 (6.18)
   Range 58.00 - 94.00 44.00 - 92.00 44.00 - 94.00

Characteristics by A4 vs LEARN 

tableby(SUBSTUDY ~ .,
  data = dd %>%
    filter(!SUBSTUDY %in% 'SF') %>% # excluding screen-fails
    select(SUBSTUDY, `A4 amyloid eligibility^b^`, `Age at screening (yrs)`, 
      `Education (yrs)`, `APOE genotype`, `Amyloid PET SUVr^b^`, Sex, Ethnicity,
      `Marital status`, `Participant retired`, PACC, MMSE, 
      `Logical memory delay`, `Digit symbol`, `FCSRT (2xFree + Cued)`), 
  digits = 2) %>%
  summary(title = "Baseline characteristics of A4-randomized^a^ and LEARN-enrolled cohorts")
Baseline characteristics of A4-randomizeda and LEARN-enrolled cohorts
A4a (N=1169) LEARN (N=538) Total (N=1707) p value
A4 amyloid eligibilityb < 0.001
   Negative 0 (0.0%) 538 (100.0%) 538 (31.5%)
   Positive 1169 (100.0%) 0 (0.0%) 1169 (68.5%)
Age at screening (yrs) < 0.001
   Mean (SD) 71.92 (4.81) 70.53 (4.32) 71.48 (4.71)
   Range 65.00 - 85.74 65.00 - 85.60 65.00 - 85.74
Education (yrs) 0.127
   Mean (SD) 16.57 (2.81) 16.79 (2.63) 16.64 (2.76)
   Range 7.00 - 30.00 8.00 - 30.00 7.00 - 30.00
APOE genotype < 0.001
   N-Miss 0 2 2
   E2/E2 2 (0.2%) 5 (0.9%) 7 (0.4%)
   E2/E3 61 (5.2%) 66 (12.3%) 127 (7.4%)
   E2/E4 35 (3.0%) 10 (1.9%) 45 (2.6%)
   E3/E3 417 (35.7%) 342 (63.8%) 759 (44.5%)
   E3/E4 560 (47.9%) 111 (20.7%) 671 (39.4%)
   E4/E4 94 (8.0%) 2 (0.4%) 96 (5.6%)
Amyloid PET SUVrb < 0.001
   Mean (SD) 1.33 (0.18) 0.99 (0.07) 1.22 (0.22)
   Range 0.97 - 2.09 0.79 - 1.16 0.79 - 2.09
Sex 0.440
   Male 475 (40.6%) 208 (38.7%) 683 (40.0%)
   Female 694 (59.4%) 330 (61.3%) 1024 (60.0%)
Ethnicity 0.820
   Hispanic or Latino 34 (2.9%) 18 (3.3%) 52 (3.0%)
   Not Hispanic or Latino 1124 (96.2%) 516 (95.9%) 1640 (96.1%)
   Unknown or Not reported 11 (0.9%) 4 (0.7%) 15 (0.9%)
Marital status 0.164
   Married 836 (71.5%) 386 (71.7%) 1222 (71.6%)
   Widowed 102 (8.7%) 52 (9.7%) 154 (9.0%)
   Divorced 170 (14.5%) 67 (12.5%) 237 (13.9%)
   Never married 42 (3.6%) 29 (5.4%) 71 (4.2%)
   Unknown/Other 19 (1.6%) 4 (0.7%) 23 (1.3%)
Participant retired 0.702
   Yes 877 (75.0%) 411 (76.4%) 1288 (75.5%)
   No 274 (23.4%) 121 (22.5%) 395 (23.1%)
   Not Applicable 18 (1.5%) 6 (1.1%) 24 (1.4%)
PACC < 0.001
   Mean (SD) -0.00 (2.68) 0.79 (2.35) 0.25 (2.61)
   Range -12.52 - 7.75 -8.70 - 6.64 -12.52 - 7.75
MMSE < 0.001
   Mean (SD) 28.78 (1.28) 29.03 (1.17) 28.86 (1.25)
   Range 22.00 - 30.00 23.00 - 30.00 22.00 - 30.00
Logical memory delay < 0.001
   Mean (SD) 12.62 (3.68) 13.54 (3.35) 12.91 (3.61)
   Range 0.00 - 23.00 3.00 - 24.00 0.00 - 24.00
Digit symbol 0.011
   Mean (SD) 48.64 (10.02) 49.97 (9.89) 49.06 (10.00)
   Range 15.00 - 86.00 0.00 - 79.00 0.00 - 86.00
FCSRT (2xFree + Cued) < 0.001
   Mean (SD) 77.35 (6.29) 78.67 (5.83) 77.77 (6.18)
   Range 44.00 - 92.00 58.00 - 94.00 44.00 - 94.00

a A4-randomized cohort (n=1169) includes other participants in addition to the modified intention-to-treat population (mITT n=1147) reported in the A4 trial (Sperling et al. 2023). The modified intention-to-treat population population that was reported for the A4 trial results include those who received at least one dose of solanezumab or placebo and underwent assessment for the primary end point. Please refer to the vignette(topic = 'A4-Primary-Results') file for code to reproduce the baseline characteristics and primary findings of the A4 study.
b Refer to A4 Amyloid PET Eligibility Methods PDF document (vignette(topic = 'imaging_PET_VA_methods')) regarding these amyloid-related measures, the eligibility determination process and the modifications made to the SUVR algorithm.

References

R Core Team. 2024. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. https://www.R-project.org/.
Sperling, Reisa A, Michael C Donohue, Rema Raman, Michael S Rafii, Keith Johnson, Colin L Masters, Christopher H van Dyck, et al. 2023. “Trial of Solanezumab in Preclinical Alzheimer’s Disease.” New England Journal of Medicine 389 (12): 1096–1107. https://doi.org/10.1056/NEJMoa2305032.
Sperling, Reisa A, Michael C Donohue, Rema Raman, Chung-Kai Sun, Roy Yaari, Karen Holdridge, Eric Siemers, Keith A Johnson, Paul S Aisen, and for the A4 Study Team. 2020. “Association of Factors with Elevated Amyloid Burden in Clinically Normal Older Individuals.” JAMA Neurology 77 (6): 735–45. https://doi.org/10.1001/jamaneurol.2020.0387.