Commentary (Wahl): Diagnostic Use of Radiolabeled Antibodies for Cancer

Since 1995 represents the 20th anniversary of Kohler and Milstein's description of the hybridoma/monoclonal antibody technology, it is an appropriate time to take stock of progress in this area. The article by Harrison and Tempero provides a useful review and update of the field of monoclonal antibody imaging in this anniversary year.

As the authors point out, the major limitation to the use of antibodies for imaging or therapy has been the low relative and absolute delivery of radioantibody to tumors. This low delivery has several causes, including the lack of truly tumor-specific antigens, the relatively low blood flow to many tumors, and the limited diffusion of large macromolecules into tumors. Thus, relatively poor antibody delivery to many types of cancer still represents the most important impediment to progress in the area of immunoglobulin targeting.

Despite this limitation, monoclonal antibodies appear to be having an increased impact on patient care in selected cases, particularly in patients with colorectal cancer. While to date only one antibody has been approved for clinical diagnostic use by the FDA, there are evolving clinical niches for which this approach is becoming a standard of practice in some centers. One clinical example is the use of antibody scanning in cases of rising serum carcinoembryonic antigen (CEA) level to detect and localize recurrence in patients with a history of colorectal cancer who have negative conventional imaging tests. Also, in patients in whom partial hepatectomy is planned for supposedly isolated metastases to the liver, the use of monoclonal antibody scanning to determine whether disease is disseminated, and thus unresectable, seems to have obvious clinical benefit. If expensive surgical procedures (like partial hepatectomy) are avoided through antibody scanning, then economics, patient preference, and changes in the planned procedure would appear to support the use of the antibody scan [1].

As is the case with any new test, a given medical center must have sufficient experience with the monoclonal antibody agent (or must seek outside review) to apply it to routine imaging, being certain that the proper position on the receiver-operator-characteristic (ROC) curve for interpretation is attained. Since false-negative results are common, selecting an operating point on the curve for interpretations where false-positive results are infrequent may be most appropriate. This is done so that an incorrect decision to not perform surgery will occur only on rare occasions [2].

Promising Results in Prostate and Lung Cancer Imaging

The authors, perhaps due to space limitations, do not mention promising initial clinical results of monoclonal antibody imaging in prostate and lung cancer. These data suggest that, in a fraction of patients, a rising prostate-specific antigen (PSA) level following treatment of prostate cancer may be due to disease recurrence remote from the prostate bed [3].

Results of imaging with monoclonal antibodies in lung cancer are encouraging in some preliminary studies as well. Perhaps non-small-cell or small-cell lung cancer would be better staged by monoclonal antibody imaging than by conventional methods [4]. It should be realized, however that very promising results are also being reported with the use of fluorine-18 2-deoxyglucose positron emission tomography (FDG PET) scanning for primary lesion detection, staging, and assessment for metastatic disease in patients with lung cancer [5]. The PET method requires more expensive imaging equipment than antibody scanning but appears to be progressing rapidly in its ability to image a diverse array of tumors of varying histologies[6].

The authors briefly describe methods to enhance targeting of antibodies to tumors, a very important goal if the method is to be extended to more diseases. One approach is regional antibody delivery via subcutaneous or intraperitoneal routes. Regional delivery approaches offer major theoretical advantages over intravenous delivery, as following regional delivery, the monoclonal antibody is presented at very high concentration to the antigen, which favors binding. Some problems associated with subcutaneous antibody de- livery, such as high targeting to normal lymph nodes, may be avoided by delayed imaging times and the use of antibody cocktails, which permit clearance of normal nodal tracer activity [7].

New Approaches to Antibody Targeting

The authors also discuss several new approaches to antibody targeting, including the possibility of pretargeting, a method still not fully explored clinically. While complementarity-deter-mining region-grafted humanized an- tibodies are expected to result in lower human antimouse antibodies levels than their murine analogs, the former will likely have longer circulation times in the blood in humans than intact antibody, and thus, higher background blood activity levels. Consequently, they may not represent a major improvement or an "answer" to the targeting problem, though they likely will reduce "human antimouse antibody" problems.

Antibody fragments and, perhaps, peptides bound to receptors represent attractive alternatives to intact antibody. While their delivery to tumors is lower than intact antibody, they clear from the blood and background rapidly, allowing for earlier imaging following tracer injection. It should also be realized that the single-chain antibodies consisting of variable portions of the heavy and light chain are potentially more immunogenic than human antibodies in some cases, due to the unnatural (and thus antigenic) peptide linkage between heavy- and light-chain areas. Nonetheless, protein engineering and peptide chemistry methods should make available new compounds for imaging or treatment.

Finally, the authors briefly discuss cancer treatment with antibodies, or radioimmunotherapy (RAIT). As they point out, the rate of biologic clearance of radioantibodies from individual patients can vary considerably, and tracer doses of antibody are commonly administered prior to antibody therapy to determine clearance rates. They caution that human antimouse antibodies could affect the clearance of the treatment dose, and question whether such a use of antibodies before RAIT will be rational.

Such concerns are appropriate, but in our experience are not justified when relatively high antibody protein doses are administered for the diagnostic and treatment doses. In such circumstances, the quantity of human antimouse antibodies is generally insufficient to measurably alter tracer targeting. Thus, in our nuclear medicine clinic, antibody tracer scans prior to RAIT now represent the most common indication for immunoscintigraphy [8].

Applications Likely to Grow

In summary, as Harrison and Tempero note, applications of antibody scanning/targeting are likely to grow as new agents are approved and as human and smaller structural analogs (possibly even peptides) become available. Increasingly, in addition to studies focusing on diagnostic accuracy, careful studies of cost-benefit will be necessary before these agents become standard components of diagnostic or treatment algorithms in most centers. Similarly, probe-guided surgery, while clearly feasible and apparently reasonably accurate in trained hands, will require careful cost-benefit analysis before its use is widely adopted.

The appropriate role of PET scanning of tumor metabolism, which may require a substantial financial outlay for equipment, relative to antibody scanning, will also become clearer as the strengths and weaknesses of each technique are better understood. Finally, the applications of antibody scanning prior to RAIT will grow, as patient-individualized RAIT doses may well be safer than treatment doses based only on patient body weight or other factors.

Thus, 20 years after Kohler and Milstein's discovery, monoclonal antibodies are assuming a gradually expanding role in clinical medical practice, with a major increase in use expected when diagnostic antibodies for prostate and lung cancer and the antibody-based radioimmunotherapies are approved.

References:

1. Corman ML, Galandiuk S, Block GE, et al: Immunoscintigraphy with 111In-satumomab pendetide in patients with colorectal adenocarcinoma: Performance and impact on clinical management. Dis Colon Rectum 37(2):129-137, 1994.

2. Wahl RL: Monoclonal antibodies in nuclear medicine, 1992, in Freeman LM (ed): Nuclear Medicine Annual 1992, pp 91-103. New York, Raven Press, 1992.

3. Kahn D, Williams RD, Seldin DW, et al: Radioimmunoscintigraphy with 111-indium labeled CYT-356 for the detection of occult prostate cancer recurrence. J Urol 152(5; pt 1):1490-1495, 1994.

4. Friedman S, Sullivan K, Salk D, et al: Staging non-small-cell carcinoma of the lung using technetium-99m-labeled monoclonal antibodies. Hematol Oncol Clin North Am 4(6):1069-1078, 1990.

5. Wahl RL, Quint LE, Greenough RL, et al: Staging of mediastinal non-small-cell lung cancer with FDG PET, CT, and fusion images: Preliminary prospective evaluation. Radiology 191:371-377, 1994.

6. Wahl RL, Hawkins R, Larson SM, et al: Proceedings of a National Cancer Institute Workshop: PET as a clinical tool in oncology-A research agenda. Radiology 193:604-606, 1994.

7. Wahl RL, Liebert M, Headington J, et al: Lymphoscintigraphy in melanoma: Initial evaluation of a low protein dose monoclonal antibody cocktail. Cancer Res 50(suppl):941-948s, 1990.

8. Kaminski MS, Zasadny KR, Francis IR, et al: Radioimmunotherapy of B-cell lymphoma with [131-I]anti-B1 (anti-CD20) antibody. N Engl J Med 329(7):459-465, 1993.