Is there an alternative to string.Replace that is case-insensitive?

Seems like string.Replace should have an overload that takes a StringComparison argument. Since it doesn't, you could try something like this:

public static string ReplaceString(string str, string oldValue, string newValue, StringComparison comparison)
{
    StringBuilder sb = new StringBuilder();

    int previousIndex = 0;
    int index = str.IndexOf(oldValue, comparison);
    while (index != -1)
    {
        sb.Append(str.Substring(previousIndex, index - previousIndex));
        sb.Append(newValue);
        index += oldValue.Length;

        previousIndex = index;
        index = str.IndexOf(oldValue, index, comparison);
    }
    sb.Append(str.Substring(previousIndex));

    return sb.ToString();
}

From MSDN
$0 - "Substitutes the last substring matched by group number number (decimal)."

In .NET Regular expressions group 0 is always the entire match. For a literal $ you need to

string value = Regex.Replace("%PolicyAmount%", "%PolicyAmount%", @"$$0", RegexOptions.IgnoreCase);

Expanding on C. Dragon 76's popular answer by making his code into an extension that overloads the default Replace method.

public static class StringExtensions
{
    public static string Replace(this string str, string oldValue, string newValue, StringComparison comparison)
    {
        StringBuilder sb = new StringBuilder();

        int previousIndex = 0;
        int index = str.IndexOf(oldValue, comparison);
        while (index != -1)
        {
            sb.Append(str.Substring(previousIndex, index - previousIndex));
            sb.Append(newValue);
            index += oldValue.Length;

            previousIndex = index;
            index = str.IndexOf(oldValue, index, comparison);
        }
        sb.Append(str.Substring(previousIndex));
        return sb.ToString();
     }
}

Based on Jeff Reddy's answer, with some optimisations and validations:

public static string Replace(string str, string oldValue, string newValue, StringComparison comparison)
{
    if (oldValue == null)
        throw new ArgumentNullException("oldValue");
    if (oldValue.Length == 0)
        throw new ArgumentException("String cannot be of zero length.", "oldValue");

    StringBuilder sb = null;

    int startIndex = 0;
    int foundIndex = str.IndexOf(oldValue, comparison);
    while (foundIndex != -1)
    {
        if (sb == null)
            sb = new StringBuilder(str.Length + (newValue != null ? Math.Max(0, 5 * (newValue.Length - oldValue.Length)) : 0));
        sb.Append(str, startIndex, foundIndex - startIndex);
        sb.Append(newValue);

        startIndex = foundIndex + oldValue.Length;
        foundIndex = str.IndexOf(oldValue, startIndex, comparison);
    }

    if (startIndex == 0)
        return str;
    sb.Append(str, startIndex, str.Length - startIndex);
    return sb.ToString();
}

Since .NET Core 2.0 or .NET Standard 2.1 respectively, this is baked into the .NET runtime [1]:

"hello world".Replace("World", "csharp", StringComparison.CurrentCultureIgnoreCase); // "hello csharp"

[1] https://docs.microsoft.com/en-us/dotnet/api/system.string.replace#System_String_Replace_System_String_System_String_System_StringComparison_

a version similar to C. Dragon's, but for if you only need a single replacement:

int n = myText.IndexOf(oldValue, System.StringComparison.InvariantCultureIgnoreCase);
if (n >= 0)
{
    myText = myText.Substring(0, n)
        + newValue
        + myText.Substring(n + oldValue.Length);
}

Regex.Replace(strInput, strToken.Replace("$", "[$]"), strReplaceWith, RegexOptions.IgnoreCase);

The regular expression method should work. However what you can also do is lower case the string from the database, lower case the %variables% you have, and then locate the positions and lengths in the lower cased string from the database. Remember, positions in a string don't change just because its lower cased.

Then using a loop that goes in reverse (its easier, if you do not you will have to keep a running count of where later points move to) remove from your non-lower cased string from the database the %variables% by their position and length and insert the replacement values.

(Since everyone is taking a shot at this). Here's my version (with null checks, and correct input and replacement escaping) ** Inspired from around the internet and other versions:

using System;
using System.Text.RegularExpressions;

public static class MyExtensions {
    public static string ReplaceIgnoreCase(this string search, string find, string replace) {
        return Regex.Replace(search ?? "", Regex.Escape(find ?? ""), (replace ?? "").Replace("$", "$$"), RegexOptions.IgnoreCase);          
    }
}

Usage:

var result = "This is a test".ReplaceIgnoreCase("IS", "was");

Let me make my case and then you can tear me to shreds if you like.

Regex is not the answer for this problem - too slow and memory hungry, relatively speaking.

StringBuilder is much better than string mangling.

Since this will be an extension method to supplement string.Replace, I believe it important to match how that works - therefore throwing exceptions for the same argument issues is important as is returning the original string if a replacement was not made.

I believe that having a StringComparison parameter is not a good idea. I did try it but the test case originally mentioned by michael-liu showed a problem:-

[TestCase("œ", "oe", "", StringComparison.InvariantCultureIgnoreCase, Result = "")]

Whilst IndexOf will match, there is a mismatch between the length of the match in the source string (1) and oldValue.Length (2). This manifested itself by causing IndexOutOfRange in some other solutions when oldValue.Length was added to the current match position and I could not find a way around this. Regex fails to match the case anyway, so I took the pragmatic solution of only using StringComparison.OrdinalIgnoreCase for my solution.

My code is similar to other answers but my twist is that I look for a match before going to the trouble of creating a StringBuilder. If none is found then a potentially large allocation is avoided. The code then becomes a do{...}while rather than a while{...}

I have done some extensive testing against other Answers and this came out fractionally faster and used slightly less memory.

    public static string ReplaceCaseInsensitive(this string str, string oldValue, string newValue)
    {
        if (str == null) throw new ArgumentNullException(nameof(str));
        if (oldValue == null) throw new ArgumentNullException(nameof(oldValue));
        if (oldValue.Length == 0) throw new ArgumentException("String cannot be of zero length.", nameof(oldValue));

        var position = str.IndexOf(oldValue, 0, StringComparison.OrdinalIgnoreCase);
        if (position == -1) return str;

        var sb = new StringBuilder(str.Length);

        var lastPosition = 0;

        do
        {
            sb.Append(str, lastPosition, position - lastPosition);

            sb.Append(newValue);

        } while ((position = str.IndexOf(oldValue, lastPosition = position + oldValue.Length, StringComparison.OrdinalIgnoreCase)) != -1);

        sb.Append(str, lastPosition, str.Length - lastPosition);

        return sb.ToString();
    }