diff --git a/src/plugins/intel_gpu/src/kernel_selector/kernels/gather/gather_kernel_ref.cpp b/src/plugins/intel_gpu/src/kernel_selector/kernels/gather/gather_kernel_ref.cpp
index acdd3bf84e5224..5f5dea544a66bb 100644
--- a/src/plugins/intel_gpu/src/kernel_selector/kernels/gather/gather_kernel_ref.cpp
+++ b/src/plugins/intel_gpu/src/kernel_selector/kernels/gather/gather_kernel_ref.cpp
@@ -155,6 +155,19 @@ static inline std::vector<std::string> GetOrder(size_t size) {
     return idx_order;
 }
 
+static inline std::vector<std::string> GetFinalIndexOrder(size_t size) {
+    std::vector<std::string> idx_order;
+
+    OPENVINO_ASSERT(size > 4, "[GPU] Only support 5 or 6 dimensions");
+
+    if (size == 5) {
+        idx_order = {"b", "f", "0", "z", "0"};
+    } else if (size == 6) {
+        idx_order = {"b", "f", "0", "w", "z", "0"};
+    }
+    return idx_order;
+}
+
 static std::string GetDictionaryIndexOrder(const gather_params& params, size_t axis) {
     auto idx_order = GetOrder(params.outputs[0].GetDims().size());
     auto input_axis_index_macro = "INPUT_AXIS_INDEX";
@@ -181,21 +194,27 @@ static std::string GetDictionaryIndexOrder(const gather_params& params, size_t a
 }
 
 static std::string GetIndicesIdxOrder(const gather_params& params, size_t axis, int64_t batch_dim) {
-    std::vector<std::string> idx_order = GetOrder(params.outputs[0].GetDims().size());
-    auto zero_val = "0";
+    std::vector<std::string> idx_order;
 
-    size_t indices_dims_num = GetNonEmptyDimsNumber(params.inputs[1]);
+    if ((axis == (size_t)batch_dim) && (axis > 1) && (params.inputs[1].GetDims().size() > 4)) {
+        idx_order = GetFinalIndexOrder(params.outputs[0].GetDims().size());
+    } else {
+        idx_order = GetOrder(params.outputs[0].GetDims().size());
+        auto zero_val = "0";
 
-    // Shift indices of Gather indices input related to output dims
-    for (size_t i = batch_dim; i < indices_dims_num; i++)
-        idx_order[i] = idx_order[axis + i - batch_dim];
+        size_t indices_dims_num = GetNonEmptyDimsNumber(params.inputs[1]);
 
-    for (size_t i = indices_dims_num; i < idx_order.size(); i++)
-        idx_order[i] = zero_val;
+        // Shift indices of Gather indices input related to output dims
+        for (size_t i = batch_dim; i < indices_dims_num; i++)
+            idx_order[i] = idx_order[axis + i - batch_dim];
 
-    // Fix size to inputs[1] dims size
-    for (size_t i = 0; i < params.outputs[0].GetDims().size() - params.inputs[1].GetDims().size(); i++)
-        idx_order.pop_back();
+        for (size_t i = indices_dims_num; i < idx_order.size(); i++)
+            idx_order[i] = zero_val;
+
+        // Fix size to inputs[1] dims size
+        for (size_t i = 0; i < params.outputs[0].GetDims().size() - params.inputs[1].GetDims().size(); i++)
+            idx_order.pop_back();
+    }
 
     return GetOrderString(idx_order);
 }
diff --git a/tests/layer_tests/tensorflow_tests/test_tf_AdjustHue.py b/tests/layer_tests/tensorflow_tests/test_tf_AdjustHue.py
index fe5e18d2b07f94..cdf77a086ed76b 100644
--- a/tests/layer_tests/tensorflow_tests/test_tf_AdjustHue.py
+++ b/tests/layer_tests/tensorflow_tests/test_tf_AdjustHue.py
@@ -50,8 +50,6 @@ def create_adjust_hue_net(self, input_shape, input_type, special_case):
     def test_adjust_hue_basic(self, input_shape, input_type, special_case,
                               ie_device, precision, ir_version, temp_dir,
                               use_legacy_frontend):
-        if ie_device == 'GPU' and (input_shape == [2, 4, 4, 3] or input_shape == [3, 4, 12, 12, 3]):
-            pytest.skip('150766: Accuracy issue on GPU')
         self._test(*self.create_adjust_hue_net(input_shape, input_type, special_case),
                    ie_device, precision, ir_version, temp_dir=temp_dir,
                    use_legacy_frontend=use_legacy_frontend)