Typical SCID and Newborn Screening for SCID and T Cell Defects

Jennifer Puck, MD

Department of Pediatrics University of California San Francisco and Benioff Children’s Hospital Smith Cardiovascular Research Institute San Francisco, California Disclosure

Spouse employed by InVitae, Inc., a clinical gene sequencing company. NBS Case: term infant, 0 TREC, 473 Actin

Low total lymphocytes

T-

B+ NK+

No naïve T cells Severe Combined Immunodeficiency, SCID, Before Newborn Screening

• Rare; many genes. • Healthy at birth, but get recurrent, severe infections, failure to thrive. • High mortality in first year. • Absent or low T cells. • Absent or non-functional B cells. • Possible cure by bone marrow transplant from a healthy matched donor (first in 1968). David Veer (1971-1984), Texas Children’s Hospital hp://www.wired.com/news/images/full/bubbleboy3_f.jpg Disclosures and Gratitude

State of CA, Newborn screening immunology expert Baxter Inc, Immunoglobulin clinical trial Research Support from: Jeﬀrey Modell Foundaon, Primary Immunodeﬁciency Center NIH NICHD, NIAID DHHS Maternal and Child Health Bureau CDC Newborn Screening Branch, Division of Lab Sciences TREC: T Cell Receptor Excision Circle, a Biomarker for Thymic T Cell Production

• Normal diverse T cell repertoire is made by cung and pasng alternate DNA secons of receptor genes • Excised DNA forms T Cell Receptor Excision Circles (TRECs) as a byproduct. • TRECs can be detected by PCR.

TCRD locus Vα V δRec Dδ Jδ Cδ ΨJα Jα Cα J Dδ δ // // // // Cδ

TCRA locus // TREC

PCR across joint

SCID Newborn Screening, 2008: Wisconsin

2008

Screening Not screening

DHHS Sec. Advis. Commiee: SCID nominated for newborn screening SCID Newborn Screening, December 2009: 3 states, Navajo Nation, Puerto Rico

2008 2009

2009

Screening 2009 Not screening CDC and JMF Grants for Pilot SCID Screening in response to Sec. Advis Commiee refusal due to no proven screening test SCID Newborn Screening, December 2010: 5 states, Navajo Nation, Puerto Rico

2008 2009 2010

2010 2009

Screening 2009 Not screening DHHS Sec. Advis. Commiee approves SCID as a disease for newborn screening. California and New York add 750,000 births/year. SCID Newborn Screening, December 2011: 6 states, Navajo Nation

2008 2009 2011 2010 2011

2010 2009

2010, Screening stopped Not screening

Louisiana, Puerto Rico stopped when pilot funds ran out SCID Newborn Screening, December 2012: 11 states, Navajo Nation

2008 2009 2011 2010 2011

2012 2010 2012

2009

2012 2012 2010, Screening stopped 2012 Not screening SCID Newborn Screening, December 2013: 17 states, Navajo Nation

2013 2008 2010 2009 2013 2011 2011 2013 2013 2013 2013 2012 2010 2012 2009

2012 2012 2010, Screening 2008-2012 stopped Screening 2013-2015 2012 Not screening 11 NBS programs pool data from 3 million infants screened. Kwan et al., JAMA 2014 SCID Newborn Screening, December 2014: 27 states, Navajo Nation

2014 2014 2013 2014 2008 2010 2009 2013 2011 2014 2013 2011 2014 2013 2014 2013 2013 2012 2014 2014 2010 2012 2014 2009

2014 2012 2012 2010, Screening 2008-2012 stopped Screening 2013-2015 2012 Not screening

US FDA approved SCID screening assay (PerkinElmer) SCID Newborn Screening, December 2015: 28 states, Navajo Nation

2014 2014 2013 2014 2008 2010 2009 2013 2011 2014 2013 2011 2014 2013 2014 2013 2013 2012 2014 2014 2010 2012 2014 2009 2015 2015 2015 2014 2012 2012 2010, Screening 2008-2012 stopped Screening 2013-2015 2012 Not screening SCID Newborn Screening, September, 2016: 41 states, Navajo Nation and Puerto Rico

2014 2016 2016 2014 2013 2014 2016 2008 2016 2016 2016 2010 2009 2013 2011 2014 2013 2011 2014 2013 2014 2013 2013 2012 2014 2014 2016 2010 2012 2014 2016 2009 2015 2015 2016 2016 2015 2014 2012 2016 2012 2010, Screening 2008-2012 stopped Screening 2013-2015 2012 2016 Screening, 2016 Not screening >85% of USA births; 93% expected by end of 2016 Inial TREC TREC >22 Normal assay, DNA TRECs copies/uL of TREC >22 ≤22 blood Posive / Repeat β-Acn TREC OK Abnormal with TREC ≤5

California β-Acn β-Acn DAF bad Regular NBS β-Acn OK TREC NICU Incomplete TREC 6-22 β-Acn bad Algorithm DAF CBC, lymphocyte subsets by NBS Program Repeat TREC ≤22 Low TREC DAF or Heel x2 at one contract Lab; Incomplete Sck interpreted by Immuno TREC >22 Normal Consultants

Not normal: <1,500 T cells or absent Normal: ≥1,500 naïve CD4/CD45RA T cells – Refer to CD3 T cells with Immunodeﬁciency Center naïve cells Inial TREC TREC >22 Normal assay, DNA TRECs copies/uL of TREC >22 ≤22 blood Posive / Repeat β-Acn TREC OK Abnormal with TREC ≤5

Not normal: <1,500 T cells or absent Normal: ≥1,500 naïve CD4/CD45RA T cells – Refer to CD3 T cells with Immunodeﬁciency Center naïve cells SCID Definition

“Classic” SCID pre-Newborn Screening: • No T cells; no specific anbody, failure to thrive, thrush, recurrent infecons. New definion with screening needs to be based on lab values: • “Typical” SCID: <300/uL autologous T cells; <10% of normal proliferaon to mitogens such as PHA; no producon of specific anbodies. Proven with null mutaon(s) in known SCID gene. • “Leaky” SCID (includes Omenn syndrome): 300-1500 T cells or more, but low to absent naïve CD45RA T cells, impaired PHA, hypomorphic mutaon(s) in SCID gene. Many SCID Genes, Distinct T, B, NK profiles

• IL2RG (common γ chain, X-linked) T- B+ NK- • JAK3 (γ chain associated Janus kinase) T- B+ NK- • IL7R (IL-7 receptor α chain) T- B+ NK+ • CD45 (membrane tyrosine kinase) T- B+ NK+ • TCRD/E/Z ( TCR CD3 δ,ε,ξ chains) T- B+ NK+ • RAG1/RAG2 T- B- NK+ • DCLRE1C (Artemis) T- B- NK+ • LIG4 (DNA ligase IV) T- B- NK+ • PRKDC (DNA PKcs) T- B- NK+ • ADA (adenosine deaminase) T- B- NK- • AK2 (reticular dysgenesis, deafness) T- B+/- NK+ • TTC7A (multiple bowel atresias) T- B+/- NK+ • RMRP (cartilage hair hypoplasia) T- B+/- NK+ • FOXN1 (nude mouse) T- B+ NK+ • CORO1A (Coronin 1A) T- B+/- NK+ Primary Immune Deficiency Treatment Consortium (PIDTC)

>40 USA, Canadian and European Centers from 2009 Uniform data collection

Retrospective Study 6902: Cross-sectional evaluation

Prospective Study 6901: Natural history, observational Identify factors affecting outcomes of HCT, ERT, or GT

5 year survival for Typical SCID depends more on infection than age at HCT

• >3.5 months with active infection worse than any other group. • <3.5 months and >3.5 months with no infection had best survival; • >3.5 months with infection resolved not p<0.001 significantly worse than <3.5 months or uninfected.

Pai et al, NEJM, 2014 371:434-46. PIDTC SCID Diagnosis by Screening vs. Infection, Family History % Diagnosed Genotypes of Typical and Leaky SCID

Reports from Transplant California, with Centers, no Screening TREC Screening Duke University, European 3½ years, 1.7 ×106 infants centers (estimates) Unknown 3% RAG2 1% DCLRE1C 5% Unknown RMRP 1% 12% TTC7A 1% IL2RG 28% JAK3 7% IL2RG JAK3 6% 50% ADA RAG1 14% ADA 12% 18% IL7R IL7R 10% 15%

RAG1 1% Overall Survival ~74% Overall Survival 94% 4 Years of California SCID Newborn Screening (2010-2014) Leaky SCID 2 million infants 3% Idiopathic Typical screened. TCL/Variant SCID SCID 12% 255 flow cytometry 4% 2nd tier tests. (1/8,000 infants) 109/255 had <1500 T Normal T Syndrome 13% cells/uL (43%). cells by flow 1/55,000 Typical and 57% Leaky SCID. Secondary 1/180,000 Idiopathic 5% T Cell Lymphopenia Preterm 6% Non-SCID Conditions: Get Early Diagnosis, Avoid Live Vaccines Mulsystem syndromes with variable T cell deﬁciency 57% DiGeorge/chromosome 22q11.2 deleon 15% Trisomy 21 3% Ataxia telangiectasia 2% CHARGE syndrome

Secondary T lymphopenia 25% Congenital cardiac anomalies 38% Other congenital anomalies 13% Vascular leakage, third spacing, hydrops 3% Neonatal leukemia 3-5% Maternal immunosuppressive medicaons

Extreme preterm birth—T cells become normal over me

Idiopathic T lymphopenia—few naïve T cells, no maternal cells, poor immune funcon, no gene defect at outset (when solved move to correct category) Search for Gene Diagnosis Using Deep Sequencing

Proband DNA Filter to retain variants with: Good sequence quality Capture Fragment, exon add linkers Changes predicted in gene fragments product

Sequence Absence in controls both ends of Locaon in gene or pathway of fragments interest, or in gene expressed in a ssue of interest

Raw data: short Presence in relaves according to end reads, expected inheritance paern; bi-direconal variants in same gene in similarly aﬀected, unrelated individuals

Align to reference sequence; List variants Damage to protein funcon Initial Clinical Findings • Leaky SCID: 0 TRECs, low T cells, B+, NK+, poor PHA • Microcephaly • Hypertelorism • Abnormal nasal creases • Neonatal tooth • Decreased tone • Desquamang rash • Hirsusm, loose skin with extra folds • Absent corpus callosum Exome Trio: De novo BCL11b het missense N441K. Normal human cord blood CD34+ differentiation after transduction with lentiviruses Empty GFP, Patient mutant Bcl11b (N441K) + GFP, or Bcl11b-siRNA + GFP

GFP GFP 25 mut 16 mut siRNA siRNA 14 * 20 12 * 10 15 * 8 10 6 * 4 % GFP+ CD19+ %GFP+ cells

5 CD3+ cells %GFP+ 2

0 0 0 5 10 15 20 25 30 0 5 10 15 20 25 30 Days in co-culture on OP9 monolayers Days in co-culture on OP9-DL1 monolayers

* p < 0.05, GFP v/s mut and GFP v/s siRNA Mut Bcl11b and Bcl11-b siRNA do not affect B cell development, but abrogate T cell development on OP9 cells. This indicates a dominant negative effect of the N441K Bcl11b. T cell phenotype: Zebraﬁsh bcl11ba-ATG-MO Knockdown in Tg(lck:EGFP), 5 dpf bcl11ba-ATG-MO Cont 4ng 2ng

Lateral view

Dorsal view

11/11 14/23 17/18 Punwani et al., NEJM in Press, 2016. Collaboraon with David Weist and Yong Zhang zebraﬁsh lab (Fox Chase, Phila), Steven Brenner group and TCS genomic analysis (Berkeley and Hyderabad). T Lymphopenia in CLOVES Syndrome

Congenital Lipomatous Overgrowth with Vascular Malformaons, Epidermal nevi and Skeletal abnormalies-PIK3CA mutaons 150 T cells/uL; loss of T cells and IgG into cysc hygromas Preterm Low Birthweight Infants with Low TRECs and T Lymphopenia

4000

3500

3000

2500 BW 300g

2000

CD3 T cells /uL CD3 T cells /uL Threshold 1500 for follow-up

BW 560g 1000 BW 700g BW 557g 500 BW 445g BW 490g

0 0 2 4 6 8 10 12 14 16 18 Age in weeks Opportunities to Learn Population Based Aspects of T cell Immune Deficiency • SCID, 1/54,000 births in CA; 1/58,000 in 11 program collaboration • All T lymphopenias, 1/23,000 births • DiGeorge, 22q11.2 deletion – Overall incidence ~1/5,000 births. – Cardiac defects, hypocalcemia, variable T cell immunodeficiency. – California DiGeorge cases with <1,500 T cells/uL represent 5% of all DiGeorge cases. • Ataxia telangiectasia, 1/200,000 births – Over half detected at birth by TREC screening. Following Infants with Low Lymphocyes, FILL CIS, USIDNet, JMF, IDF Collaboration Infants are being recognized who have low lymphocytes. In addition to SCID and leaky SCID, newborn screening finds non-SCID conditions with low TRECs and low T cells. The FILL project aims to • capture the spectrum of infants born with low lymphocytes; • record how these infants are managed; and • learn their course and outcomes over the first 18 months of life.

Idiopathic T Lymphopenia (Variant SCID)

• Persistently low T cells and TRECs, low naïve CD45RA T cells, no maternal engraftment. • No known SCID gene mutation. • Impaired T cell and/or antibody responses. • When an etiology is found, case is moved to the appropriate category. What Immune Disorders Is the TREC Test Missing? • Infants with SCID who are not screened or followed up • Defect post-TCR recombination; diverse T cells with functional problems: MHC-II, CD40L, ZAP70, etc • SCID gene defect leaky enough to allow TRECs to be made • Syndrome with variable T cell deficiency, for cases with enough T cell generation to have TRECs above cutoff • Maternal neutralization of a metabolic defect (ADA), making TRECs normal at birth, reduced later • PID not involving T cells or not evident at birth: XLA, CVID, CGD Conclusions 1. SCID comprises treatable, serious, genetic immune deficiencies affecting 1/50,000-1/60,000 births. 2. Early diagnosis permits optimal treatment, outcomes. 3. Population based newborn screening with the TREC test identifies pre-symptomatic SCID, caused by a wide range of gene defects, known and unknown. 4. TREC screening also detects non-SCID conditions with low T cells, offering clinical benefit and opportunities to study the spectrum of disorders. 5. A high index of suspicion and new testing approaches are needed for primary immune defects not picked up by TREC screening.

Thanks to Many Collaborators

California Dept. of Pub Health Support Bob Currier and colleagues NIH Perkin Elmer NCRR UCSF CTSI Quest Nichols Lab NIAID USIDNet and RO3

UCSF Primary Immune Mort Cowan Deficiency Treatment Christopher Dvorak Consortium NBSTRN Stanford Matt Porteus Jeffrey Modell Foundation Sean McGhee CA Dept. of Pub Health LA Children’s Hospital Center for Disease Control Joseph Church and Prevention Neena Kapoor DHHS Maternal and Child

UCLA Health Bureau Don Kohn, E. Richard Stiehm