diff --git a/sklearn/metrics/ranking.py b/sklearn/metrics/ranking.py
index aa2e5425976e9..6d023cd7b9b37 100644
--- a/sklearn/metrics/ranking.py
+++ b/sklearn/metrics/ranking.py
@@ -23,6 +23,7 @@
 import numpy as np
 from scipy.sparse import csr_matrix
 from scipy.stats import rankdata
+from functools import partial
 
 from ..utils import assert_all_finite
 from ..utils import check_consistent_length
@@ -188,8 +189,8 @@ def _binary_uninterpolated_average_precision(
                                  sample_weight=sample_weight)
 
 
-
-def roc_auc_score(y_true, y_score, average="macro", sample_weight=None):
+def roc_auc_score(y_true, y_score, average="macro", sample_weight=None,
+                  pos_label=None):
     """Compute Area Under the Curve (AUC) from prediction scores
 
     Note: this implementation is restricted to the binary classification task
@@ -226,6 +227,12 @@ def roc_auc_score(y_true, y_score, average="macro", sample_weight=None):
     sample_weight : array-like of shape = [n_samples], optional
         Sample weights.
 
+    pos_label : int or str, optional
+        Default is ``None``. If ``None``, pos_label is considered to be 1.
+
+        The label of the positive class. For multilabel-indicator y_true,
+        pos_label is fixed to 1.
+
     Returns
     -------
     auc : float
@@ -251,18 +258,33 @@ def roc_auc_score(y_true, y_score, average="macro", sample_weight=None):
     0.75
 
     """
-    def _binary_roc_auc_score(y_true, y_score, sample_weight=None):
+    def _binary_roc_auc_score(y_true, y_score, sample_weight=None,
+                              pos_label=None):
         if len(np.unique(y_true)) != 2:
             raise ValueError("Only one class present in y_true. ROC AUC score "
                              "is not defined in that case.")
 
         fpr, tpr, tresholds = roc_curve(y_true, y_score,
+                                        pos_label=pos_label,
                                         sample_weight=sample_weight)
         return auc(fpr, tpr, reorder=True)
 
-    return _average_binary_score(
-        _binary_roc_auc_score, y_true, y_score, average,
-        sample_weight=sample_weight)
+    y_type = type_of_target(y_true)
+    if y_type == "binary":
+        _partial_binary_roc_auc_score = partial(_binary_roc_auc_score,
+                                                pos_label=pos_label)
+        return _average_binary_score(
+            _partial_binary_roc_auc_score, y_true, y_score, average,
+            sample_weight=sample_weight)
+    else:
+        if pos_label is not None and pos_label != 1:
+            raise ValueError("Parameter pos_label is fixed to 1 for"
+                             "multilabel-indicator y_true. Do not set "
+                             "pos_label or set pos_label to either None "
+                             "or 1.")
+        return _average_binary_score(
+            _binary_roc_auc_score, y_true, y_score, average,
+            sample_weight=sample_weight)
 
 
 def _binary_clf_curve(y_true, y_score, pos_label=None, sample_weight=None):
diff --git a/sklearn/metrics/tests/test_common.py b/sklearn/metrics/tests/test_common.py
index 5f775aaf9ac8f..0eb7873be92d3 100644
--- a/sklearn/metrics/tests/test_common.py
+++ b/sklearn/metrics/tests/test_common.py
@@ -257,6 +257,7 @@
 
     "macro_f0.5_score", "macro_f1_score", "macro_f2_score",
     "macro_precision_score", "macro_recall_score",
+    "roc_auc_score",
 ]
 
 # Metrics with a "labels" argument
@@ -595,7 +596,7 @@ def test_invariance_string_vs_numbers_labels():
 
     for name, metric in THRESHOLDED_METRICS.items():
         if name in ("log_loss", "hinge_loss", "unnormalized_log_loss",
-                    "brier_score_loss"):
+                    "brier_score_loss", "roc_auc_score"):
             # Ugly, but handle case with a pos_label and label
             metric_str = metric
             if name in METRICS_WITH_POS_LABEL:
@@ -607,7 +608,7 @@ def test_invariance_string_vs_numbers_labels():
                                err_msg="{0} failed string vs number "
                                        "invariance test".format(name))
 
-            measure_with_strobj = metric(y1_str.astype('O'), y2)
+            measure_with_strobj = metric_str(y1_str.astype('O'), y2)
             assert_array_equal(measure_with_number, measure_with_strobj,
                                err_msg="{0} failed string object vs number "
                                        "invariance test".format(name))
diff --git a/sklearn/metrics/tests/test_ranking.py b/sklearn/metrics/tests/test_ranking.py
index db80691663606..277846bac8b3f 100644
--- a/sklearn/metrics/tests/test_ranking.py
+++ b/sklearn/metrics/tests/test_ranking.py
@@ -371,6 +371,19 @@ def test_roc_curve_drop_intermediate():
                               [1.0, 0.9, 0.7, 0.6, 0.])
 
 
+def test_roc_auc_score_pos_label_multilabel_indicator():
+    # Raise an error for multilabel-indicator y_true with
+    # pos_label other than None or 1
+    y_true = np.array([[1, 0], [0, 1], [0, 1], [1, 0]])
+    y_pred = np.array([[0.9, 0.1], [0.1, 0.9], [0.8, 0.2], [0.2, 0.8]])
+    roc_auc_score_1 = roc_auc_score(y_true, y_pred, pos_label=None)
+    assert_almost_equal(roc_auc_score_1, 0.75)
+    roc_auc_score_2 = roc_auc_score(y_true, y_pred, pos_label=1)
+    assert_almost_equal(roc_auc_score_2, 0.75)
+    assert_raises(ValueError, roc_auc_score, y_true, y_pred,
+                  pos_label=0)
+
+
 def test_auc():
     # Test Area Under Curve (AUC) computation
     x = [0, 1]