diff --git a/src/rdkafka_cgrp.c b/src/rdkafka_cgrp.c
index 397ddf5250..210132fb7f 100644
--- a/src/rdkafka_cgrp.c
+++ b/src/rdkafka_cgrp.c
@@ -5299,12 +5299,61 @@ rd_kafka_cgrp_modify_subscription(rd_kafka_cgrp_t *rkcg,
         return RD_KAFKA_RESP_ERR_NO_ERROR;
 }
 
+/**
+ * Remove existing topic subscription (KIP 848).
+ */
+static rd_kafka_resp_err_t
+rd_kafka_cgrp_consumer_unsubscribe(rd_kafka_cgrp_t *rkcg,
+                                   rd_bool_t leave_group) {
+
+        rd_kafka_dbg(rkcg->rkcg_rk, CGRP, "UNSUBSCRIBE",
+                     "Group \"%.*s\": unsubscribe from current %ssubscription "
+                     "of size %d (leave group=%s, has joined=%s, %s, "
+                     "join-state %s)",
+                     RD_KAFKAP_STR_PR(rkcg->rkcg_group_id),
+                     rkcg->rkcg_subscription ? "" : "unset ",
+                     rkcg->rkcg_subscription ? rkcg->rkcg_subscription->cnt : 0,
+                     RD_STR_ToF(leave_group),
+                     RD_STR_ToF(RD_KAFKA_CGRP_HAS_JOINED(rkcg)),
+                     rkcg->rkcg_member_id ? rkcg->rkcg_member_id->str : "n/a",
+                     rd_kafka_cgrp_join_state_names[rkcg->rkcg_join_state]);
+
+        rd_kafka_timer_stop(&rkcg->rkcg_rk->rk_timers,
+                            &rkcg->rkcg_max_poll_interval_tmr, 1 /*lock*/);
+
+        if (rkcg->rkcg_subscription) {
+                rd_kafka_topic_partition_list_destroy(rkcg->rkcg_subscription);
+                rd_kafka_cgrp_subscription_set(rkcg, NULL);
+        }
+
+        /*
+         * Clean-up group leader duties, if any.
+         */
+        rd_kafka_cgrp_group_leader_reset(rkcg, "unsubscribe");
+
+        if (leave_group)
+                rkcg->rkcg_flags |= RD_KAFKA_CGRP_F_LEAVE_ON_UNASSIGN_DONE;
+
+        /* FIXME: Why are we only revoking if !assignment_lost ? */
+        if (!rd_kafka_cgrp_assignment_is_lost(rkcg))
+                rd_kafka_cgrp_revoke_all_rejoin(rkcg, rd_false /*not lost*/,
+                                                rd_true /*initiating*/,
+                                                "unsubscribe");
+
+        rkcg->rkcg_flags &= ~(RD_KAFKA_CGRP_F_SUBSCRIPTION |
+                              RD_KAFKA_CGRP_F_WILDCARD_SUBSCRIPTION);
+
+        return RD_KAFKA_RESP_ERR_NO_ERROR;
+}
 
 /**
  * Remove existing topic subscription.
  */
 static rd_kafka_resp_err_t rd_kafka_cgrp_unsubscribe(rd_kafka_cgrp_t *rkcg,
                                                      rd_bool_t leave_group) {
+        if (rkcg->rkcg_group_protocol == RD_KAFKA_GROUP_PROTOCOL_CONSUMER) {
+                return rd_kafka_cgrp_consumer_unsubscribe(rkcg, leave_group);
+        }
 
         rd_kafka_dbg(rkcg->rkcg_rk, CGRP, "UNSUBSCRIBE",
                      "Group \"%.*s\": unsubscribe from current %ssubscription "
@@ -5326,8 +5375,7 @@ static rd_kafka_resp_err_t rd_kafka_cgrp_unsubscribe(rd_kafka_cgrp_t *rkcg,
                 rd_kafka_cgrp_subscription_set(rkcg, NULL);
         }
 
-        if (rkcg->rkcg_group_protocol == RD_KAFKA_GROUP_PROTOCOL_CLASSIC)
-                rd_kafka_cgrp_update_subscribed_topics(rkcg, NULL);
+        rd_kafka_cgrp_update_subscribed_topics(rkcg, NULL);
 
         /*
          * Clean-up group leader duties, if any.