What is documented, and what it cannot tell us.
Four Wichita Falls children were diagnosed with brain tumors between August 2017 and January 2018. Three of the four were high-grade gliomas. Two were glioblastoma.
This page sets out that information, the geographic and timing details connected to it, and — just as importantly — the reasons those details cannot establish a cause. It is here so the circumstances are on the record in a form qualified researchers and public-health professionals can evaluate.
Children are named here only where their family has given permission. Where no permission exists, the diagnosis is described without identifying details. No street addresses or birth dates are published, and nothing on this page comes from a medical record obtained without the family's consent. The one clinical detail published about an individual child is my own son's tumour pathology and genetic testing, which I have chosen to publish as his mother.
Four children, between August 2017 and January 2018.
These are the four Wichita Falls pediatric brain tumor diagnoses I am aware of between August 2017 and January 2018. Three of the four fall within about five months of each other; the fourth is documented only to the year 2017. This is not a registry count and it is not a complete accounting — it is what has been reported to me directly by families or is otherwise publicly known. There may be more.
Gunner Smith
Age 10. Glioblastoma multiforme.
A fourth Wichita Falls child
A pediatric brain tumor diagnosis within the same period. The family has not consented to website publication, so the date and all identifying details are withheld.
Shahreen Alam
Age 5. Diffuse intrinsic pontine glioma (DIPG).
Dominic Watson
Age 11. Glioblastoma multiforme with giant cell features, WHO Grade 4. Diagnosed January 20, 2018.
Gunner and Dominic were both diagnosed with glioblastoma — Dominic's with giant cell features. Shahreen was diagnosed with diffuse intrinsic pontine glioma. All three are high-grade gliomas. The two glioblastomas are WHO grade 4 by definition, and DIPG is grade 4 under the current classification, as diffuse midline glioma, H3 K27-altered.
Two of the four children were diagnosed with glioblastoma. That is a more specific observation than "four brain tumors," and it is the reason the pattern is documented here rather than left unrecorded. It is still not evidence of a shared cause.
In 2021 the World Health Organization revised its classification of central nervous system tumors. Under the current system the term "glioblastoma" is no longer applied to children. Pediatric high-grade gliomas are classified as separate entities, including diffuse midline glioma H3 K27-altered, which is where DIPG now falls and which is WHO grade 4 by definition.
The diagnoses these children received in 2017 and 2018 would
likely carry different names today. That matters here for two
reasons. Published incidence figures from before 2021 may not
map cleanly onto current categories, and any comparison across
that change should be made carefully.
Source: WHO Classification of Tumours of the Central Nervous
System, 5th edition (2021).
Dominic's tumor showed p53 expression on pathology. Genetic testing did not identify a mutation in his TP53 gene, and did not identify an inherited cancer predisposition syndrome.
That is evidence against a heritable predisposition in his case alone. It says nothing about the other three children, and it does not point toward any other explanation. Most childhood cancers arise from changes that are neither inherited nor attributable to any identified exposure.
Four diagnoses do not create a scientifically valid probability calculation. The tumors are not all the same type, the children's exposure histories differ, and the geographic and time boundaries around these cases were drawn after the diagnoses were known rather than defined in advance — which is exactly the condition under which clusters can appear in data by chance.
Evaluating whether this pattern is unusual requires population denominators, registry-verified case counts, exposure histories and statistical methods appropriate to post-hoc cluster review. That is work for public-health professionals, not for a website.
What a family can do is preserve the record — who, where, when, and what diagnosis — so the information exists when someone qualified looks at it.
These are rare tumors.
Pediatric high-grade brain tumors are uncommon. Glioblastoma and DIPG are both rare in children, and published national incidence figures reflect that. Rarity alone does not make a group of cases a cluster, but it is the reason a group of cases gets noticed. The figures below predate the 2021 reclassification described above, so they describe the categories as they stood when these children were diagnosed.
| Tumor type | Population | Reported incidence | Source |
|---|---|---|---|
|
Glioblastoma Gunner's and Dominic's diagnoses |
United States Ages 0–19 |
0.18 per 100,000 per year 95% confidence interval 0.16–0.19 |
Grochans et al. 2022, citing Ostrom et al. [1] |
|
Diffuse intrinsic pontine glioma (DIPG) Shahreen's diagnosis |
United States Median age 6–7 years |
Approximately 150–300 children
diagnosed per year Nationwide count, not a population rate |
DIPG Registry, via DIPG.org [2] |
|
Brain and other CNS tumors, all types For comparison |
United States Ages 0–19 |
2.99 per 100,000 per year In 2017, up from 2.13 in 1975 |
Hossain et al. 2021, SEER 1975–2017 [3] |
Sources.
[1] Grochans S, Cybulska AM, Simińska D, Korbecki J, Kojder K,
Chlubek D, Baranowska-Bosiacka I. Epidemiology of Glioblastoma
Multiforme — Literature Review. Cancers.
2022;14(10):2412. doi:10.3390/cancers14102412. The 0.18 per 100,000
age-adjusted rate for ages 0–19 is reported there from Ostrom
et al.
Grochans et al. also cite a figure of 0.85 per 100,000 for ages
0–18 drawn from a different group of sources. The 0.18 figure
used here is the age-adjusted SEER/CBTRUS rate reported from Ostrom
et al., which is the more conservative of the two and the more
directly comparable to the other rows in this table.
[2] DIPG statistics, DIPG.org (The Cure Starts Now Cancer Resource
Network), citing the DIPG Registry; figures stated as current as of
February 7, 2019. This is an advocacy organization reporting
registry data, not a peer-reviewed publication.
[3] Hossain MJ, Xiao W, Tayeb M, Khan S. Epidemiology and
prognostic factors of pediatric brain tumor survival in the US:
Evidence from four decades of population data. Cancer
Epidemiology. 2021;72:101942.
doi:10.1016/j.canep.2021.101942.
How to read these numbers. A rate of 0.18 per
100,000 means that in a population of 100,000 children, fewer than
one glioblastoma diagnosis would be expected in a year. These
figures describe national averages over multi-year periods. They
are not a basis for calculating how likely it was that four
diagnoses occurred in Wichita Falls within this window, for the
reasons set out below.
Two of the four families lived close together.
Gunner and Dominic lived approximately 400 feet apart in the same small area of southwest Wichita Falls. Their families' periods of residence there overlapped by roughly 16 months.
Street addresses are not published. Distance from a treatment plant is not published either, and it would not mean what a reader might assume: everyone on a distribution system receives the same treated water, and NDMA in a chloraminated system tends to increase with time in the pipes rather than with closeness to the plant.
Residence period and the reuse period
The shared period of residence ran from approximately February 2014 through June 2015.
Wichita Falls' emergency direct potable reuse system operated from July 2014 through July 2015 — approximately twelve months, overlapping most of that shared period.
Shared residence and shared timing establish that these children were in the same place during the same period. They do not establish what either child was exposed to, at what concentration, or for how long. Individual exposure was never measured and cannot be reconstructed now.
View the full water timeline →NDMA is a recognized carcinogen. That is a separate question from these diagnoses.
The scientific literature on NDMA is well developed. What it supports, and what it does not support, should be stated precisely.
Classification
NDMA is classified as probably carcinogenic to humans (IARC Group 2A) and as a probable human carcinogen by EPA, based primarily on animal studies.
Target organs in animals
In animal studies, NDMA principally causes liver tumors, with lung and kidney tumors also reported. It is not primarily associated with brain tumors in that literature.
N-nitroso compounds and brain tumors
Research on N-nitroso compounds generally — a broader chemical family that includes NDMA — has examined possible associations with childhood brain tumors, including through prenatal exposure. The findings are mixed and not conclusive.
What this means here
NDMA in drinking water is a legitimate public-health concern in its own right. Nothing in the literature establishes that NDMA caused any of these four diagnoses, and this site does not claim that it did.
Everything this page cannot do.
These limits are not caveats added at the end. They are the honest description of what a documented set of circumstances can and cannot support.
Four cases is far too small a number to support statistical inference about cause. Rare events cluster by chance more often than intuition suggests.
The geographic area and the time window were identified after the diagnoses were known. Drawing a boundary around known cases makes almost any group of cases look concentrated.
Two of the four children were diagnosed with glioblastoma and three were high-grade gliomas, but the four diagnoses are not identical. Different tumor types can have different causes, which complicates the case for a single shared explanation.
NDMA was not routinely measured during the 2014–2015 direct potable reuse period. No one can state what concentrations any of these children consumed, or whether they consumed tap water at all.
The interval between a carcinogenic exposure and a tumor diagnosis is generally not short. Whether the timing here is biologically plausible is a question for qualified researchers, not an assumption to build on.
These cases are reported to me by families or publicly known. They have not been verified against the Texas Cancer Registry, and there may be additional cases I am unaware of.
A proper evaluation of whether this constitutes a statistically unusual cluster is the role of the Texas Department of State Health Services and the Texas Cancer Registry. Requesting that evaluation is one of the things this site asks for.
Because no one else is holding these pieces together.
Families know where their children lived, when they moved, when symptoms began and what they were diagnosed with. Those histories may never appear together in any government database, and they become harder to reconstruct every year.
Documenting them now is not the same as drawing a conclusion from them. It preserves information so that public-health professionals have something to evaluate if and when they choose to look.
Did your child live in the Wichita Falls area and develop cancer?
I want to hear from families whose child lived in Wichita Falls or the surrounding area and was diagnosed with cancer — any type of cancer, and whether or not you believe water or any other environmental exposure had anything to do with it.
Information about where your child lived, approximately when, their age and diagnosis may help preserve histories that otherwise exist only within individual families.
Contacting me does not mean your child will be identified on this website. Nothing about your family is published without your permission. Tell me what you are comfortable sharing, and whether I may contact you for more detail.
Contact Me →The water record is documented separately.
The water-quality findings on this site stand on their own, and do not depend on any conclusion about these diagnoses.