The pharmacokinetic fate of an immunotoxin: HAMA, HARA, ding, ding, dong

Fans of Tom Waits may recognize the cadence of our title line. His couplet continues ‘…sure didn't last too long.’ [1]. He is talking about a failed romance. Our Commentary addresses the pharmacokinetic (PK) fate of the immunotoxin anti-B4-blocked ricin (B4bR), as reported by Furman et al. in the current issue of Leukemia and Lymphoma. [2] Human anti-mouse antibody (HAMA) and anti-ricin (HARA) responses are predictable events after exposure to foreign proteins such as murine antibodies or the castor bean toxin ricin. Partly to circumvent the HAMA response, therapeutic monoclonal antibodies (mAbs) have been engineered to be chimeric, like rituximab [3], or humanized. However, anti-chimeric (HACA) responses can occur [4], and even fully humanized mAbs can be immunogenic [5]. Moreover, murine mAbs are still used, notably anti-CD20 radioimmunotherapeutic mAbs.

Patients with lymphoma are less prone to the development of HAMA than other patients with cancer, since the lymphoma itself compromises immunity. Prior treatment, too, dampens the response; in trials with 131-I tositumomab, HAMA was less common in previously treated patients than in patients with de novo disease [6,7]. But the Furman report demonstrates clearly that HAMA and HARA occur in the majority of their patients, and they cite literature which shows that antibody responses also occur after exposure to other immunotoxin mAbs, and another immunotoxin that is not a mAb (denileukin diftitox).

Does the development of HAMA or HARA compromise the efficacy of the therapeutic immunotoxin? Furman et al. speculate that it probably does not. But their pharmacokinetics (PK) data clearly show that there were lower mean serum levels of B4bR after the second course than after the first. A more critical analysis would have included an area under the curve (AUC) calculation instead of just levels, and a PK analysis of patients with HAMA/HARA versus those without. Nonetheless, we think it is legitimate to speculate that more protracted therapy with B4bR would likely have been associated with still lower levels, and consequent diminishing therapeutic returns. Might there be other reasons for the disappointing results seen in the Furman report? Certainly, a lack of direct cytotoxic activity against target cells could be due to penetration and other factors, as discussed in the article. Intrinsic or acquired resistance [8–11], and the presence of soluble factors in serum capable of suppressing ricin-based constructs, have all been observed [12].

Overall, these are encouraging times for immunotoxin therapy. In B-cell malignancies, a Pseudomonas exotoxin (PE)-based anti-CD22 immunotoxin has demonstrated excellent results in the clinic, with an 86% complete response (CR) rate in a phase I trial [13]. Another antibody drug conjugate (ADC) targeting CD22 and containing calicheamycin has shown good efficacy in B-cell lymphoma [14]; and in CD30-positive T-cell lymphoma and Hodgkin lymphoma, an anti-CD30 mAb linked to monomethyl auristatin E (an antitubulin agent) appears to be very promising [15,16]. However, resistance to mAb therapy can develop, through many possible mechanisms [17]. Neutralizing antibodies to the therapeutic mAb may be an understudied possibility. PK monitoring has provided numerous insights into the early course of patients' therapy with mAbs. There is less PK information available for patients who receive repeat or protracted mAb therapy. As more ADCs become available, and as more protracted mAb therapy is pursued, the importance of PK monitoring should be kept in mind.

Potential conflict of interest: Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.