Page 5099: in the Introduction, the data cutoff date of the final analyses from the ALCANZA study should be 28 September 2018, not 2019.

Page 5100: in Table 1, the footnote symbol † next to “Median PFS per IRF, months (95% CI)” should be next to “64.4 (50.7-75.2)” under “Brentuximab vedotin (n = 64).” The corrected Table 1 is shown below:

Page 5102: under “Patient responses by baseline disease stage or extracutaneous involvement,” in the sentence that begins “The superiority of brentuximab vedotin over physician’s choice,” “stage IVB, 29% vs not available” should read “stage IVB, 29% vs not applicable.”

Page 5103: under “Peripheral neuropathy,” in the sentence that begins “Of the 44 patients with PN,” the percentage of the 9 patients permanently discontinuing brentuximab vedotin treatment should be 20%, not 14%.

Page 5104: under “Discussion,” in the second paragraph of the right column, in the sentence that begins “Limiting the assessment of OS according to assigned therapy,” the percentage of patients in the physician’s choice arm who received brentuximab vedotin as a subsequent treatment should be 69%, not 62%.

In supplemental Table 1, under “Overall (N=128),” in row “T1,” “5/13 (16)” should read “5/31 (16).” In supplemental Table 4, under “CR,” in row “T3,” “1(7)” should read “1(17).” The updated Supplement File 1 is included in the online version of this erratum. These errors do not impact the main findings or conclusions of the article.