Selectize.js doesn't read HTML entities correctly - php

Question

I'm working with Selectize.js currently and am stuck on a problem relating to HTML entities not displaying correctly in Selectize fields, and therefore failing checks after submission. We use PHP on the backend and run every Selectize option through htmlspecialchars($option);. We then export the PHP array to a JSON array in preparation for adding it to the Selectize script/field.
//Array of options
$options = getOptions();
$optionsArray = array();
foreach($options as $item) {
$optionsArray[] = array('text' => htmlspecialchars($item), 'value' => htmlspecialchars($item));
}
//Create html script to run selectize initialization
$htmlcode = "<script>";
...
$htmlcode .= "var optionList = " . json_encode($"optionsArray") . ";";
$htmlcode .= "$('#selector').selectize({ options: optionList });";
...
$htmlcode .= "</script>";
The point here is to protect against malicious entries by encoding. This typically wouldn't be a problem, as browsers will detect the entities and convert them automatically (ie. will detect & and display as &. However, that is NOT happening inside the selectize. As you can see in the image below, outputting an option with an ampersand displays correctly on the page, but it doesn't display accurately in the Selectize field.
Example:
The problem then becomes larger, as we are comparing the selected item against the database after submission, to ensure that the selected entry actually exists. If the entry testOption&stuff actually exists, the form submission will check testOption&stuff against it, and will obviously fail the check.
//For simplicity, checking against static option
var $selectedOption = $_POST["options"];
if ($selectedOption == "testOption&stuff") {
//Do success
} else {
//Do failure
}
How can I resolve this and make it so that the Selectize field will store and display the correct entry (ie. &)?
Thanks!