How are many to many relationships in Cloud Spanner used using Spring Boot without Hibernate

Question

The Cloud Spanner Library for Java Spring Boot states the following:

While one-to-one and many-to-many relationships can be implemented in Cloud Spanner and Spring Data Cloud Spanner using constructs of interleaved parent-child tables, only the parent-child relationship is natively supported.

However I could not find any solution on how to actually implement this many-to-many relationship in my project.

With the Hibernate Dialect the @ManyToMany annotation would solve this problem. But how does someone implement it purely with the Google Spanner Data Library?

If I use the @Interleaved annotation on both entities that have the relationship, the join table needs to have the @PrimaryKey in the same order as those entities. But because both entities have the Primary Key on the keyOrder = 1, one of them is in the wrong order

com.google.cloud.spring.data.spanner.core.mapping.SpannerDataException: The child primary key column (JoinTable.ID2) at position 1 does not match that of its parent (Entity2.ID2)

@Table
public class Entity1 {

    @PrimaryKey(keyOrder = 1)
    private UUID ID1;

    @Interleaved
    private List<JoinTable> entities2;

}

@Table
public class Entity2 {

    @PrimaryKey(keyOrder = 1)
    private UUID ID2;

    @Interleaved
    private List<JoinTable> entities1;

}

@Table
public class JoinTable {

    @PrimaryKey(keyOrder = 1)
    private UUID ID1;
    @PrimaryKey(keyOrder = 2)
    private UUID ID2;

}