High-Dose Interleukin-2

David McDermott, from Beth Israel Deaconess Medical Center, presented the results of the Target trial, which attempted to determine, prospectively, whether so-called good pathologic features in the patient population could predict for a higher response rate to IL-2 than for patients with poor pathologic features. The investigators wanted to answer the following questions: what is the best target population for IL-2; does the carbonic anhydrase 9  biomarker  predict success with the drug; can we improve the success rate of IL-2. 

This trial enrolled 120 patients at 14 cancer centers from November 2006 to July 2009. It restricted participation to only those patients who had no prior therapy for mRCC (metastatic kidney cancer) of whom most had clear cell pathology, although a few patients with papillary and chromophobe histologies were also included.  Dr. McDermott did not report on adverse side effects experience by patients on the trial, however, there were two treatment-related deaths. The characteristics of the patients on the trial, based on two risk category models for prognosis of survival, were:

NB: For an explanation of ECOG category, see http://www.ackc.org/kidney-cancer-information/ecog-status/

The response results were as follows:

In addition to the above results for partial and complete responses, it should be noted that, overall, 40% of patients had some tumor shrinkage[1].  The response rates were higher than the standard historical rates for HD IL-2 (23% CR + PR), however, the median progression-free survival was only 4.2 months. The higher response rate was probably due to better patient selection, i.e., fewer co-morbidities. Of the 120 patients, only 5 were non-clear cell, none of whom responded to IL-2; the clear cell patients had a 30% response rate. 

Prior to this trial, the high staining of the biomarker CA-9 (carbonic anhydrase 9) had been considered to be a predictor of success for mRCC patients who are treated with IL-2. For this trial, the investigators divided the patients into two groups, Good and Poor Risk, based on the combined score of CA-9 staining (>85%) and a pathology risk groups assessment that was developed at Harvard for prediction of response to IL-2 treatment. The model is diagrammed as follows, followed by the tabular representation.


Response rate by CA-9 staining and Histologic Risk Group is given in the following table. To the investigators’ surprise, patients in the low-staining CA-9 group had a higher response rate, 38%, than the patients in the high-staining group, 23%. Even the combined score yielded similar results. These results contradicted the prevailing opinion among RCC oncologists that high CA-9 staining leads to better response to IL-2 treatment. Even the Histologic Risk Group yielded mixed results with Poor risk patients faring better than expected.

RCC oncologists who use IL-2 are still searching for a predictive biomarker for IL-2 response as well as looking at other targeted immunotherapies and agents like MDX-1106, which has shown to give durable results with a lower toxicity than IL-2. However, until other agents are developed, McDermott stated that, despite its limited efficacy and high toxicity, “IL-2 based immunotherapy is the only approach that can produce a response duration curve showing a median of 23.3 months” (duration of response for those patients who had a complete or partial response only).  Therefore, IL-2 is still being recommended by many RCC oncologists for first-line therapy in patients with clear cell histology only, and with good performance status. It is not recommended as a second-line therapy following targeted therapies due to toxicity issues, especially cardiac.

Novartis recently contracted with Prometheus Lab to distribute IL-2 (Proleukin) in the U.S. We spoke with a Prometheus rep after McDermott’s presentation, and he told us that Prometheus was planning to initiate a marketing campaign for the drug. Only 20% of RCC patients have ever had a conversation with an oncologist about the availability of HD IL2, and only 2-3% of patients have actually requested it from their oncologist. There are no hard numbers of how many patients with mRCC take IL-2 or even which U.S. medical centers offer it. Prometheus is conducting a survey to find the answers.

[1] Note that PR means shrinkage of at least 30% in target lesions’ diameter, PD is 20% or higher increase in diameter, and CR is complete disappearance of all target lesions – see http://imaging.cancer.gov/clinicaltrials/imaging  for fuller explanation.