BACKGROUND%20Factors%20influencing%20time%20to%20HIV-1%20seroconversion%20are%20not%20well%20established.%20Rare%20case%20reports%20describe%20patients%20with%20prolonged%20periods%20prior%20to%20HIV-1%20antibody%20generation.%20Such%20patients%20may%20have%20a%20rapidly%20progressive%20course%20but%20it%20is%20not%20known

1
Time to HIV-1 Seroconversion Similar Among
Patients with Acute HIV-1 Infection but there
are Exceptions Susan Morpeth1, Nathan Thielman1,
Julia Giner1, Prema Menezes2, Susan Fiscus2,
Georgia Tomaras1, Christopher Pilcher2, Jeffrey
Lennox3, Joseph Eron2, Charles Hicks1 1. Duke
University Medical Center, Durham, NC, USA, 2.
University of North Carolina, Chapel Hill, NC,
USA, 3. Emory University, Atlanta, GA, USA.

• Susan Morpeth MBChB
• Duke University Medical Center, Durham, NC 27710
• morpe001_at_mc.duke.edu
• Fax [phone]
• Phone [phone]

389
Case Study a 49 year old man with acute HIV-1
infection
ABSTRACT
Characteristics Significantly Associated with a
Negative Initial HIV-1 ELISA at Time of Diagnosis
of Acute HIV-1 Infection
Timeline
BACKGROUND Factors influencing time to HIV-1
seroconversion are not well established. Rare
case reports describe patients with prolonged
periods prior to HIV-1 antibody generation. Such
patients may have a rapidly progressive course
but it is not known if this is due to increased
viral virulence or host factors. METHODS The
Duke-UNC-Emory Acute HIV Research Consortium
studies adult patients with symptomatic acute
HIV-1 infection diagnosed within 30 days of
seroconversion. This prospective cohort enrolled
72 subjects from 1998-2005. Associations were
sought between initial ELISA status and
demographic variables, viral load (VL), nadir CD4
count and time to seroconversion using Wilcoxon
rank-sum and Fishers exact tests. We report one
case in which seroconversion took gt1 year to
occur. RESULTS At the time of acute HIV-1
diagnosis, 44 of 72 patients had a negative
ELISA. Median time (25-75th quartiles) from
symptom onset to first ELISA test negative ELISA
9 (5-15) days reactive ELISA 25 (16-37)
days, plt0.0001. Median initial VL was higher
among negative ELISA patients compared to
reactive ELISA patients (5,649,238 vs. 310, 525
copies/mL, plt0.0001). Patients with a negative
ELISA were more likely to be non-Caucasian
(p0.0286). Nadir CD4 counts were comparable
(medians 393, 433, pns) as was time to first
positive ELISA (medians 26 days, 25 days, pns).
A greater proportion of cases who were ELISA
negative at diagnosis were identified in 2004-5
than in the preceeding six years. A 22 yo man
presented with acute HIV-1 infection (VLgt10
million copies/mL) and had repeatedly negative
HIV-1 ELISA tests over the ensuing year, during
which he had PCP. He seroconverted only after
initiation of effective antiretroviral therapy.
Antibody responses to prior tetanus and
diphtheria immunization were normal. His HLA
type was not one associated with abnormal
responses (HLA-B72). Plasma-derived virus (at
initial diagnosis) used both CCR5 and CXCR4
co-receptors. CONCLUSIONS Subjects diagnosed with
acute HIV infection while ELISA-negative had a
significantly higher VL than those who had
already seroconverted, but nadir CD4 counts did
not differ. Median time to a reactive HIV-1 ELISA
from onset of symptoms also did not differ. In
2004-5 more cases were identified through the NC
State program, meaning they were ELISA negative
and HIV RNA PCR positive. One patient who rapidly
progressed to AIDS did not seroconvert for a full
year, despite evidence of functional B cell
responses to non-HIV antigens. Virus targeting of
HIV-1 specific T cells that augment specific B
cell responses may explain this observation.
(n number of subjects with data available) Initial ELISA Negative (n44) Initial ELISA Negative (n44) Initial ELISA Positive (n28) Initial ELISA Positive (n28) p-value
(n number of subjects with data available) Median (Interquartile range) Median (Interquartile range) p-value
HIV-1 RNA at initial ELISA (copies/mL) n37 5,649,238 (2,322,724 - 16,866,713) 310,525 (140,538 1,682,818) lt0.0001
Days from symptoms to initial ELISA n72 9 (5 15) 25 (16 37) lt0.0001
Days from symptoms until CART initiated n63 26 (19 36) 38 (23 54) 0.0148
Number (Percent) Number (Percent)
Non-white ethnicity n72 25 (57) 8 (29) 0.0286
Characteristics Not Significantly Associated with
a Negative Initial HIV-1 ELISA at Time of
Diagnosis of Acute HIV-1 Infection
HIV-ab HIV-1 antibody by Vironostika HIV-1
ELISA, bioMerieux, Durham, NC
(n number of subjects with data available) Initial ELISA Negative (n44) Initial ELISA Negative (n44) Initial ELISA Positive (n28) Initial ELISA Positive (n28) p-value
(n number of subjects with data available) Median (Interquartile range) Median (Interquartile range) p-value
Age (years) n72 29 (23 36) 32 (23 45) 0.2745
CD4 T-lymphocyte nadir recorded in 8 weeks from symptom onset (cells/mm3) n46 393 (266 589) 433 (365 656) 0.3542
Days from symptoms to positive ELISA n63 26 (20 34) 25 (17 40) 0.8724
Number (Percent) Number (Percent)
Male gender n72 38 (86) 26 (93) 0.4705

• Phenosense GTTM in April 2004
• Replicative capacity 15
• Clade B
• Resistance mutations M184M/I/V, K103N, V108V/I,
  L33V (K103N and L33V were later found to have
  been present in a pre-treatment sample from
  December 2003)
• Other assays performed
• Adequate levels of IgG, IgA, IgM, IgE, Diphtheria
  and Tetanus antibodies in April 2004
• HIV-1 antibody testing performed using
  Vironostika ELISA kit from bioMerieux, Durham, NC
  and confirmed as negative from May and June 2004
  samples using Genetic SystemsTM by Bio-Rad,
  Redmond, WA
• HIV-1 Western Blots (Bio-Rad) negative or
  indeterminate until seroconversion in September
  2004
• HLA-B72 (not known to be associated with HIV-1
  disease progression)

Co-receptor Use of Plasma-derived Virus from Case
Study Subject in March 2004
BACKGROUND

• Time to HIV-1 seroconversion is established at lt1
  month in modern cohorts and is dependent on ELISA
  technology used.
• Typical time from HIV-1 infection to AIDS is
  about 12 years but there are rapid progressors
  and slow progressors.
• A high viral load set point and low CD4 nadir is
  well known to correlate with poor prognosis but
  replicative capacity of wild-type virus,
  drug-resistant mutations and co-receptor use are
  all viral characteristics that are likely to come
  into play.
• A broad and strong cytotoxic T lymphocyte (CTL)
  response results in prompt virological control
  and a good antibody response, although there will
  be eventual viral escape in most people. A poor
  CTL response may be associated with a poor
  antibody response and slow to no control of
  virological replication and CD4 destruction.
• There have been case reports of patients with
  rapidly progressive HIV disease with seronegative
  infection by ELISA and Western blot, some of whom
  have seroconverted only after virological control
  by antiretroviral therapy.

CART combination antiretroviral therapy
CONCLUSIONS AND DISCUSSION
Subject
R5 Control
X4 Control

• Subjects diagnosed with acute HIV-1 infection
  while ELISA-negative had a significantly higher
  viral load at time of first evaluation than those
  who had already seroconverted, but nadir CD4
  counts did not differ.
• Median time to a reactive HIV-1 ELISA from onset
  of symptoms in the cohort also did not differ by
  initial ELISA status, indicating that initially
  seronegative subjects did not have a longer
  window period.
• Persons of non-white ethnicity were more likely
  to be diagnosed while still HIV-antibody
  negative, perhaps because the diagnosis of HIV
  was considered earlier.
• The increasing proportion of cases diagnosed
  while still ELISA negative in 2004-5 was likely a
  consequence of increasing numbers of cases of
  acute HIV-1 in NC being identified through the
  STAT program which teams an older ELISA assay
  with HIV RNA PCR to accomplish this goal.
• Our case study subject rapidly progressed to AIDS
  and did not seroconvert for a full year, despite
  evidence of functional B cell responses to
  non-HIV antigens. We speculate that virus
  targeting of HIV-1 specific T cells that augment
  specific B cell responses may explain this rapid
  progression.
• The case study subject had dual tropic virus in
  March 2004. If he had dual tropic virus
  initially this may partly explain his propensity
  for rapid progression, or rapid progression and
  immune collapse may be why his virus was able to
  become dual tropic early on.
• Virological control with effective CART may
  facilitate seroconversion in such cases. Could
  seroconversion be a marker of improved cytotoxic
  T lymphocyte responses?

RESULTS
Demographics of a Cohort of 72 Acutely HIV-1
Infected Subjects in North Carolina
METHODS
Median Interquartile range
Age (years) 30 23 - 41
Number Percent
Male gender 64 89
Non-white ethnicity 33 46

• We describe a patient with primary HIV-1
  infection who was persistently seronegative for
  12 months, seroconverting only after achieving
  virological control with effective antiretroviral
  therapy.
• His virus was characterized by subtype,
  replicative capacity, genotype and phenosense
  assay by Monogram Biosciences, South San
  Francisco, CA.
• Plasma derived virus isolates from the subject
  were titrated on U87 cell lines stably expressing
  CD4 alone, CD4 and CCR5, or CD4 and CXCR4 to
  assess their ability to use CCR5 and/or CXCR4 for
  productive infection.
• The case was compared to our cohort of patients
  from the Duke-UNC-Emory Acute HIV Consortium
  database, consisting of patients referred to HIV
  clinic before or within 30 days of seroconversion
  from January 1998 until December 2005.
• ELISAs used were Vironostika by bioMerieux,
  Durham, NC, Genetic Systems TM by Bio-Rad,
  Redmond, WA, Abbott HIVAB-1 and 1/2 by Abbott
  Laboratories, IL. Subjects referred from the
  Screening and Tracing Active Transmission (STAT)
  program (pooled HIV-1 RNA PCR testing of
  ELISA-negative samples from NC State testing
  sites) all had ELISAs done using the Vironostika
  HIV-1 immunoassay.
• Associations were sought between initial ELISA
  status and demographic variables, viral load
  (VL), nadir CD4 count, time to seroconversion and
  year of diagnosis using Wilcoxon rank-sum and
  Fishers exact tests in JMP 5.1, SAS Institute,
  Cary, NC.

Number of Cases of Acute HIV-1 Diagnosed while
ELISA-Negative or Positive by Year of Diagnosis
Year of initial ELISA 1998 1999 2000 2001 2002 2003 2004 2005 Total cases
Initial ELISA negative 0 2 2 5 3 3 13 16 44
Initial ELISA positive 4 7 3 4 1 1 3 5 28
Total cases 4 9 5 9 4 4 16 21 72
The number of cases of acute HIV-1 diagnosed while still ELISA-negative was significantly greater in 2004-5 (29/37, 78) than in the preceeding 6 years (15/35, 43), p0.0034. The number of cases of acute HIV-1 diagnosed while still ELISA-negative was significantly greater in 2004-5 (29/37, 78) than in the preceeding 6 years (15/35, 43), p0.0034. The number of cases of acute HIV-1 diagnosed while still ELISA-negative was significantly greater in 2004-5 (29/37, 78) than in the preceeding 6 years (15/35, 43), p0.0034. The number of cases of acute HIV-1 diagnosed while still ELISA-negative was significantly greater in 2004-5 (29/37, 78) than in the preceeding 6 years (15/35, 43), p0.0034. The number of cases of acute HIV-1 diagnosed while still ELISA-negative was significantly greater in 2004-5 (29/37, 78) than in the preceeding 6 years (15/35, 43), p0.0034. The number of cases of acute HIV-1 diagnosed while still ELISA-negative was significantly greater in 2004-5 (29/37, 78) than in the preceeding 6 years (15/35, 43), p0.0034. The number of cases of acute HIV-1 diagnosed while still ELISA-negative was significantly greater in 2004-5 (29/37, 78) than in the preceeding 6 years (15/35, 43), p0.0034. The number of cases of acute HIV-1 diagnosed while still ELISA-negative was significantly greater in 2004-5 (29/37, 78) than in the preceeding 6 years (15/35, 43), p0.0034. The number of cases of acute HIV-1 diagnosed while still ELISA-negative was significantly greater in 2004-5 (29/37, 78) than in the preceeding 6 years (15/35, 43), p0.0034. The number of cases of acute HIV-1 diagnosed while still ELISA-negative was significantly greater in 2004-5 (29/37, 78) than in the preceeding 6 years (15/35, 43), p0.0034.

• LIMITATIONS
• More frequent monitoring of HIV-1 ELISA, HIV-1
  RNA and CD4 counts than is practical in a
  clinical setting would lead to more precise
  results.
• Unfortunately peripheral blood mononuclear cells
  were not available from the case subject to
  assess his HIV-specific cytotoxic T lymphocyte
  responses.
• The tropism assay shown is from March 2004, after
  3 months of ineffective antiretroviral therapy
  (ART) and is being repeated on plasma from
  December 2003, prior to any exposure to ART.

ACKNOWLEDGEMENTS The authors gratefully
acknowledge the assistance of Monogram
Biosciences with the genotypic and phenotypic
assays of the HIV-1 virus isolated from the case
study subject. The assistance of JoAnn Kuruc and
Kara McGee with data compilation from the
Duke-UNC-Emory Acute HIV Consortium cohort is
also greatly appreciated.