Priorities for Research in HIV and Infant Feeding from the WHO Breastfeeding Meeting October 2527, 2

1
Priorities for Research in HIV and Infant
Feedingfrom the WHO Breastfeeding Meeting
October 25-27, 2006
Lynne M. Mofenson, M.D. Pediatric, Adolescent and
Maternal AIDS Branch National Institute of Child
Health and Human Development National Institutes
of Health
2
Overview

• Rationale for the WHO meeting new research
  findings related to early weaning
• Meeting recommendations for research related to
  breast milk HIV transmission
• Pathogenesis
• Risk/protection factors
• Infant feeding pattern/weaning
• ARV issues
• Prevention ARV, vaccine, modified milk
• Research on maternal decision making, counseling,
  program implementation

3
Research on Effects ofEarly Weaning on Growth,
Gastroenteritis Morbidity and Mortality
4
BOTSWANAMASHI STUDY INFANT FEEDING
COMPONENTBreastfeeding Infant AZT
Prophylaxisvs Formula Feeding
5
Mashi Trial, Botswana Infant Feeding Trial
Component Thior I et al. JAMA 2006296794-805
AZT Backbone
Effect of BF with Infant Prophylaxis
34 wk
oral
N591
N588
Formula feed 1 Month AZT
Breastfeed 6 Months AZT

• 93 never breastfed
• 95 AZT 1 mo adherence

• 18 exclusive breastfed
• 82 mixed/partial BF 1st 5
• months
• 84 AZT 6 mo adherence
• Median duration breast
• feeding, 5.9 months

Maternal characteristics similar Median CD4 366
18 CD4 lt200 Median RNA 4.4 log copies/mL

6
HIV Infection is Higher in Breastfed than
Formula-Fed Infants Despite 6 Months of AZT
Thior I et al. JAMA 2006296794-805
P0.02
Breastfeeding AZT
Formula
7
Early Mortality (Through Age 7-9 Months) Higher
in Formula-Fed than Breastfed AZT Infants
Thior I et al. JAMA 2006296794-805
overall P0.21
7 Month Difference p0.003
Formula
Breastfeeding AZT
8
No Difference in 18-Month HIV-Free Survival
Between Formula-Fed and Breastfed AZT Infants
Thior I et al. JAMA 2006296794-805
P0.48
Breastfeeding AZT
Formula
9
At 18 Months, More Breastfed Infants Infected
But More Formula-Fed Infants Died Thior I et
al. JAMA 2006296794-805
Predominant causes infant death Diarrheal
disease and pneumonia
10
Formula-Feeding Associated with Higher Rates of
Severe Pneumonia, Diarrhea and Infant Mortality
than Breastfeeding Mashi Study, Botswana
Lockman S et al. 2006 Internat AIDS Conf,
Toronto, Canada, Abs. TuPe0357
All Children
11
Formula-Feeding Associated with Higher Rates of
Severe Pneumonia and Infant Mortality in
HIV-Infected ChildrenMashi Study,
BotswanaLockman S et al. 2006 Internat AIDS
Conf, Toronto, Canada, Abs. TuPe0357
HIV-Infected Children
12
Formula-Feeding Associated with Higher Rates of
Severe Diarrhea, Wasting, Infant Mortality in
HIV-Uninfected ChildrenMashi Study,
BotswanaLockman S et al. 2006 Internat AIDS
Conf, Toronto, Canada, Abs. TuPe0357
HIV-Uninfected Children
13
KENYAKisumu Breastfeeding Study (KiBS)vs
Vertical Transmission Study (VT)Gastroenteritis
Hospitalizations and Growth Faltering
Thomas T et al. CROI 2007 in press
14
Early Weaning and Hospitalizations/Growth

• Comparison of gastroenteritis hospitalizations
  and growth in HIV-exposed uninfected infants from
  an ongoing and historical study in Kisumu, Kenya.
• KiBS Ongoing clinical trial of maternal HAART
  for prevention postnatal transmission
• Early weaning (6 months) promoted
• VT Vertical transmission study in same clinics
  1996-2001
• Traditional breastfeeding gt12 month,
  complimentary foods at 3 months

15
Rates of Gastroenteritis Hospitalizations by
Infant Age, Comparing KiBS with Early Weaning to
Natural History VT Study in Kisumu, Kenya
Age in months
( Mary Glenn Fowler MD, in press CROI 2007)
Age of Weaning in KiBs
16
Growth Faltering Post Weaning at 6 Months in KiBS
Study (N63) Compared to VT Study Without Early
Weaning (N440), Kisumu, Kenya
(Mary Glenn Fowler MD)
17
MALAWINVAZ Clinical TrialBreastfeeding and
Maternal and Infant Mortality
18
Breastfeeding by HIV-Infected Mothers and
Maternal and Infant Mortality MalawiTaha TE et
al. Bull WHO 200684546-54

• Longitudinal analysis of 2,000 mothers and their
  infants from NVAZ PMTCT clinical trial in
  conducted in Blantyre, Malawi between April
  2000-March 2003, with follow-up for 24 months.
• In the 2 years post birth, death occurred in
• 44 women (2.2)
• 310 children (15.5)
• Median duration of breastfeeding
• Overall 15 months (IQR 9-23 months)
• Exclusive 2 months (IQR 2-3 months)
• Mixed 12 months (IQR 6-18 months)

19
Breastfeeding by HIV-Infected Mothers and
Maternal and Infant Mortality MalawiTaha TE et
al. Bull WHO 200684546-54

• Breastfeeding pattern was not associated with
  maternal mortality or morbidity after adjusting
  for maternal viral load, age, hemoglobin and body
  mass index.
• Breastfeeding was associated with significantly
  reduced mortality among all infants and children,
  including both HIV-infected as well as
  HIV-uninfected children.

20
Breastfeeding Associated with Decreased Risk of
Infant Mortality Through Age 2 Years in
HIV-Uninfected and Infected Children in Malawi
Taha TE et al. Bull WHO 200684546-54
21
MALAWIPEPI vs NVAZ Clinical TrialBreastfeeding
and Gastroenteritis-Associated Infant Mortality
Kafulafula G et al. CROI 2007 in press
22
Gastroenteritis and Mortality in Malawi PEPI vs
NVAZ Trials Kafulafula G et al. CROI 2007 in
press

• Comparison of gastroenteritis frequency and
  mortality pre-/post-weaning in infants in Malawi
  enrolled in two clinical trials
• PEPI study of extended infant ARV prophylaxis
  with early weaning at 3-6 months of age
• median duration of breastfeeding 183 days.
• NVAZ study (SD NVP vs SD NVP 1 wk AZT)
  conducted in same clinics where early weaning was
  not recommended
• median duration of breastfeeding 732 days.

23
Overall GE-Related Hospitalization Between Ages 6
and 12 Months is Higher in PEPI than NVAZ
24
Mortality Among HIV-Uninfected Babies is Higher
in PEPI Study After Age 6-9 Months
PEPI
NVAZ
25
GE-Related Mortality Among HIV-Uninfected is
Higher After Age 6 Months in PEPI than NVAZ
PEPI
NVAZ
26
UGANDAMaternal HAART vs SD NVP
StudyBreastfeeding and Infant Mortalityin
Women Receiving ARV Prophylaxis or HAART
27
MTCT and Infant Mortality Among HIV-Infected
Breastfeeding Women in UgandaHomsy J et al.
2006 Internat AIDS Conf, Toronto, Canada, Abs.
TuPe0354

• MTCT and infant mortality in infants born to
  HIV-infected Ugandan mothers getting
• Maternal HAART if CD4 lt250 or WHO stage 3/4
• SD NVP (no AZT) if mother not qualify for HAART
• Counsel to exclusive BF and wean at 3-6 mos.
• 80 mothers delivered 85 infants
• 57 infants born to 52 mothers on HAART
• 91 breastfed, median 3 months
• MTCT 2 but 25 infant mortality
• 28 infants born to 28 mothers SD NVP
• 89 breastfed, median 5 months
• MTCT 18 11 infant mortality

28
MTCT Predictors Among Women Receiving HAART or
SD NVP, UgandaHomsy J et al. 2006 Internat AIDS
Conf, Toronto, Canada, Abs. TuPe0354
29
Infant Mortality Predictors in Women Receiving
HAART or SD NVP, UgandaHomsy J et al. 2006
Internat AIDS Conf, Toronto, Canada, Abs. TuPe0354
Longer duration of BF protective against infant
death even in women on HAART
30
COTE DIVOIREShort Course ARV SD NVP and
Breastfeeding vs Formula FeedingBreastfeeding
and Infant Morbidity and Mortality
31
Morbidity and Mortality in Breast- and
Formula-Fed Infants, Cote dIvoire (ANRS
1201/1202)Becquet R et al. 2006 Internat AIDS
Conf, Toronto, Canada, Abs. TuPe0350

• ANRS 1201/1202 Women received short course AZT
  SD NVP or AZT/3TC SD NVP after counseling,
  self-choice breast or formula feed infant (free
  formula).
• 557 live-born children
• 295 (53) formula-fed
• 262 (47) breastfed (median duration 4 months)
• Compared morbidity (diarrhea, respiratory
  infection or malnutrition) and severe
  morbidity/mortality (hospitalization or death) by
  infant feeding modality.
• Adjusted for infant HIV status, maternal
  education, housing, water supply, baseline
  maternal CD4, living with partner, study site and
  birth weight.

32
No Difference in Risk of Diarrhea/Respiratory
Infection or Malnutrition in Breast- vs
Formula-Fed Infants, Cote dIvoire Becquet R et
al. 2006 Internat AIDS Conf, Toronto, Canada,
Abs. TuPe0350
----- Breastfed Children ----- Formula fed
Children
33
No Difference in Risk of Hospitalization or Death
in Breast- vs Formula-Fed Infants, Cote dIvoire
Becquet R et al. 2006 Internat AIDS Conf,
Toronto, Canada, Abs. TuPe0350
----- Breastfed Children ----- Formula fed
Children
34
Morbidity and Mortality in Breast- and
Formula-Fed Infants, Cote dIvoire (ANRS
1201/1202)Becquet R et al. 2006 Internat AIDS
Conf, Toronto, Canada, Abs. TuPe0350

• The two year rates of adverse health outcome,
  hospitalization or death were similar among
  short-term breastfed and formula fed children.
• Mortality rates did not differ significantly
  between these two groups and after adjustment for
  infant HIV status were similar to long-term
  breastfed infants in earlier Ditrame study.
• Given appropriate counseling and care, access to
  clean water, and free supply of breastmilk
  substitutes, these alternatives to prolonged
  breastfeeding were safe interventions for PMTCT
  in this setting.

35
Key Research Priorities on HIV and Infant
FeedingWHO, October 2006
36
Research on the Pathogenesis of Breast Milk HIV
Transmission

• Proportion of transmission occurring via
  cell-free vs cell-associated and does this vary
  over lactation association with MTCT
• Immune content/quality of BM of HIV mothers
  (such as antibody levels) important component
  of breastfeeding value lies in the protective
  effect of breast milk for the infant against
  infectious diseases
• HIV breastfeeding women in general
• HIV breastfeeding women who are sick
• Superinfection does it occur, if yes, frequency
  and risk factors

37
HAART Reduces Breast Milk HIV-1 Cell-Free But
Not HIV-1 Cell-Associated Viral Load Shapiro R
et al. 12th Retrovirus Conf, Boston 2005 (Abs.
793b)

• Evaluated effect of HAART on suppressing breast
  milk HIV RNA or DNA.
• Study nested in MASHI, comparing breastfeeding
  women on HAART to a group of women with
  comparable HIV disease stage who did not receive
  ART.
• 23 (88) of 26 women on HAART had undetectable (lt
  50 copies/ml) HIV RNA in breast milk, compared
  with 9 (36) of 25 who did not receive HAART (p
  0.0001).
• However, there was no difference in proportions
  of women with undetectable HIV DNA in breast milk.

38
Cell-Associated (HIV DNA) but not Cell-Free Virus
(HIV RNA) Associated with Risk of Postnatal MTCT
Kenya Rousseau CM et al. J Infect Dis
20041901880-8
Analysis among 134 mother-infant pairs with
infants who were PCR negative at birth who
subsequently were HIV PCR positive in Nairobi,
Kenya
39
Postnatal MTCT Associated with Breast Milk HIV
DNA Regardless of Age but with RNA Only Age gt9
MonthsTanzania Koulinska IN et al. JAIDS
20064193-99
Adjusted for maternal CD4 count at delivery and
HIV disease stage at baseline.
40
Research on Risk Factors and Protective Factors
for Breast Milk HIV Transmission

• Risk factors for BF transmission
• Quantify transmission risk around time of weaning
  and risk factors
• Primary maternal HIV acquisition during BF
• Extent of problem
• Effect on postnatal transmission
• Implications
• Early vs late transmission different factors?
• Risk factors in the ARV era (see later)
• Viral factors such as viral clade, tropism and
  selective transmission
• Evaluate and better understand protective factors
  (e.g., alpha defensins, SLPI, CCR5 Ab, exclusive
  BF)

41
BM Viral Load Increases After Early Rapid Weaning
Infants at Increased Transmission Risk if
Resume Breastfeeding? Thea D et al. AIDS
2006201539-47

• BM virus detectable _at_ gt 50 copies/mL data for
  those with detectable virus only
• No increase in BM viral load seen in women who
  continued to BF

42
Early Rapid Breastfeeding Cessation ( _at_ 4 Mos
PP) Increases Risk of Breast Health Problems and
Reduces Duration of Amenorrhea - Zambia
(ZEBS)Thea D et al. AIDS 2006201539-47
Note The prevalence of breast health problems
among mothers who stopped BF is comparable to
other studies (HIV, non-HIV)
43
Research Related to Infant Feeding Patterns and
Weaning

• Early weaning (lt6 months)
• Safety for infant (infections, malnutrition,
  mortality, HIV-free survival meta-analysis?)
• Optimal timing (can we better define AFASS)
• Effect of weaning/type weaning on maternal milk
  (e.g., confirm abrupt weaning increases BM HIV)
• Optimal duration of transition from BF
• Feasibility/effectiveness of different
  interventions to optimize nutrition and protect
  against health/nutritional risks of early weaning

44
Research Related to Infant Feeding Patterns and
Weaning

• Weaning after 6 months
• Risks of transition from exclusive BF to BF with
  complementary feeding at gt6 months
• Infant morbidity and mortality
• HIV transmission
• Optimal timing of weaning (can we better define
  AFASS at older age)
• Infant safety of early weaning before 12-24 mos
• Feasibility/effectiveness of different
  interventions to optimize nutrition in older
  infant when weans

45
Research Related to Antiretroviral Treatment of
Breastfeeding Women

• Breastfeeding and HIV transmission in ARV era in
  countries where AFASS not met
• Effect of ARV on timing of transmission
• ARV treatment vs short IU/IP ART prophylaxis
• Risk factors for BM transmission in women on ARV
  treatment (e.g., CD4 association)
• ARV penetration in BM
• ARV levels in BF infant
• Infant safety of ARV exposure through BM
• Efficacy in reducing postnatal MTCT
• Effect ARV on virus in BM and development of
  resistant virus in BM
• Resistance in infants who become infected while
  BF

46
Prevention Research Antiretroviral Prophylaxis

• Maternal HAART for prophylaxis
• Maternal safety including safety of stopping
  HAART after prolonged period (re SMART study)
• Infant safety
• Efficacy for reducing BF MTCT optimal regimen
• Infant ARV prophylaxis
• Is it safe and does it work if yes, how
  effective
• Optimal prophylaxis duration and regimen
• First 6 months (6 wks, 14 wks, 6 mos?)
• After 6 months if AFASS not met? Prevention of
  late (age gt6 mos) BF transmission
• Questions for both maternal/ prophylaxis
• Early weaning issues (how long prophylaxis, when
  wean, complications of early weaning)
• Risk/cost/benefit

47
Prevention Research Immunization

• Passive (PACTG 185 done, HPTN 027 ongoing)
• Active HIV vaccine
• Critical to evaluate If could cover early BF
  period with short temporary ARV prophylaxis while
  inducing immunity with vaccine would allow
  prolonged safe breastfeeding
• Immunogenicity, safety phase I (ongoing
  canarypox, planned Merck adenovirus vaccine)
• Efficacy (long-term)
• Combined with passive or short infant ARV
• Safety of vaccine/effect disease in HIV-infected
  children (inadvertent administration if IU/IP
  infection undetected at time first immunization
  or becomes infected while getting vaccine)

48
Prevention Research Alternatives to
Replacement Feeding

• Treatment of milk
• Heat treatment (HTBM)
• Microbicide
• Effect on cell-free and cell-associated virus
• Effect on milk components
• Safety for infant (microbicide)
• Efficacy, feasibility, effectiveness
• Feasibility of time-limited use of treatment if
  breast pathology or infant oral pathology, during
  weaning
• Alternative methods (re-lactation, milk bank, wet
  nurse)
• HIV infection risk from a breastfeeding infected
  infant to uninfected wet nurse

49
Research on Counseling, Program Implementation
and Monitoring

• Factors influencing maternal decision making on
  infant feeding acceptability, feasibility,
  partner involvement, stigma, psychological
  stressors
• Counseling assessment of quality what
  determines effective counseling training
• Implementation of infant feeding counseling
• Cultural context
• Different mechanisms/counselors (eg, lay)
• Community-based interventions
• Compare effectiveness different interventions
• Promotion of EBF

50
Suggestions for Addressing Research Priorities
51
Some Mechanism to Address Research Priorities

• Combine data from existing/ongoing studies (eg,
  ZVitambo, ZEBS,VTS, BHITS, Ditrame, other
  observational studies)
• Address infant feeding pattern,
  morbidity/mortality, risk factors (eg, CD4)
• Await results from ARV clinical trials to
  determine where to move next related to ARV
  prophylaxis
• Critical to use samples/data from these studies
  to address pathogenesis, risk issues, ARV
  questions
• Need long-term follow-up mothers/infants (e.g.,
  endpoint is not infection rate at 12 months)

52
Some Mechanism to Address Research Priorities

• New trials needed
• Phase III trial of short ARV prophylaxis HIV
  vaccine in HIV-exposed neonates
• Results of ARV prophylaxis trials will affect
  study design
• Trial will need to be large, international
• Trial need to be collaborative as opposed to
  competing trials
• How to handle infants who need to continue to BF
  after age 6 months because dont meet AFASS
• Could be several complimentary studies
• Continue breastfeed? What is HIV risk?
• Continue/start ARV prophylaxis?
• Treatment of expressed milk? Feasibility?
• Wean with specific nutritional and health
  interventions for weaned infants?