Phase II trial of the regulatory T cell-depleting agent, denileukin diftitox, in patients with unresectable stage IV melanoma

This study was a single arm, open label phase II study of DAB/IL2 undertaken from 2007 to 2010 at the James Graham Brown Cancer Center, University of Louisville, Louisville, Kentucky. The primary objective was to determine the response rate of DAB/IL2 in stage IV metastatic melanoma patients. A secondary objective was the determination of overall survival after DAB/IL2 administration. The clinical trial registration number is NCT00299689 (at Clinicaltrials.gov).

This clinical trial was approved by the University of Louisville Human Subjects Committee. Only patients with distant metastases from cutaneous, ocular, mucosal melanoma or melanoma of unknown primary were eligible for inclusion. All patients fulfilled the following criteria: (i) primary tumor must have been documented by histopathologic analysis; (ii) metastatic disease must have been documented by radiologic examinations (CT scan or PET scan); and (iii) disease recurrences occurring greater than 5 years after the original diagnosis must have been biopsy proven. Written informed consent was obtained from each patient prior to enrollment and the trial was conducted in accordance with the Declaration of Helsinki.

All patients were subjected to fusion FDG-PET/CT or CT imaging within 1 month prior to receiving the first dose of DAB/IL2 and within 1 month after receiving the last dose of DAB/IL2. DAB/IL2 (Eisai) was purchased through third party payers and was administered as follows: 12 μg/kg, IV over 30 min every 24 h for 4 doses (cycles repeated every 21 days; 1-4 cycles). All patients had renal function tests, blood counts, and a complete physical examination prior to each cycle of DAB/IL2. The endpoint definitions were determined from qualitative radiological assessments performed by board-certified radiologists after two cycles using the following criteria:

Mixed response (MR): Significant (> 30%) tumor regressions simultaneous with significant growth (> 20%) in individual tumors (see Figure 1B for an example).

Adverse events were collected by reviewing the physician dictations and nursing notes during and 1 month following the last administration of DAB/IL2.

We administered four daily doses of DAB/IL2 (12 μg/kg; 2-4 21 day cycles) to a total of 60 stage IV melanoma patients. The vast majority of patients enrolled in the study had metastatic melanoma involving distant organs (stage M1C; 70%) and the most commonly affected organs were the lung (47%) and liver (47%) (Table 1). 82% of patients had been treated with at least one prior systemic regimen and the majority were treated with two or more prior systemic therapies. The most common previous treatment regimens included biochemotherapy (vinblastine, dacarbazine, cisplatin, IL-2 and interferon alpha) and high dose IL-2 (Table 1). The most common adverse events reported were nausea (38%), fatigue (21%), emesis (16%), rash (15%) and chills (10%) (Table 2) and these side effects can be easily managed with symptomatic as opposed to immunosuppressive agents. Interestingly, 5% of patients reported pain associated with their tumors which may reflect inflammation caused by DAB/IL2. In this trial, only one patient developed an autoimmune disorder, vitiligo, as a result of DAB/IL2 administration. We suspect that this case of clinically insignificant vitiligo likely resulted from immune cross-reactivity against antigens expressed by both melanoma cells and melanocytes.

We observed several examples of partial and mixed responses which are typical of immunotherapeutic agents. For example, an 82 year-old male developed multiple hepatic metastases (red box, second panel; Figure 1A) and a large duodenal mass (red arrow, first panel, Figure 1A) which caused significant nausea, vomiting and weight loss. After four cycles of DAB/IL2, he experienced the complete regression of his hepatic metastases confirmed by FDG-PET imaging and resolution of his symptoms but only a modest reduction in his duodenal mass (compare Baseline to +4 Months, Figure 1A; the increased ¹⁸ F-FDG uptake in the left kidney is due to hydronephrosis which is unrelated to melanoma). Next, an 83 year-old male received three cycles of DAB/IL2 and experienced marked regression of a large subcutaneous mass, a pelvic mass (Figure 1B, see bottom 2 horizontal red arrows just above the bladder [which normally contains tracer]) and a peritoneal mass (Figure 1B, right vertical arrow). Simultaneously, a large conglomeration of left axillary masses expanded (Figure 1B, dashed circle), paratracheal lymph nodes worsened (Figure 1B, upper arrows) and a peritoneal mass appeared and expanded with treatment (Figure 1B, left vertical arrow). This is a typical clinical example of a mixed response to DAB/IL2. A 78 year-old female experienced a dramatic reduction in metastases involving the liver, lung and bone that has persisted for 15 months with the exception of a single small right paratracheal lymph node (Figure 2A). A 47 year-old male who had previously progressed through high dose IL-2, biochemotherapy and several experimental agents also had a marked global reduction in hepatic, lung and subcutaneous metastatic burden (Figure 2B). As a final clinical example, a 62 year-old male who progressed after receiving anti-CTLA4 and experienced debilitating right upper quadrant pain, nausea/vomiting and fatigue associated with widespread hepatic metastases experienced a substantial partial response that was durable for at least 15 months (Figure 2C). These examples of partial but durable clinical responses to DAB/IL2 are suggestive of an immunotherapeutic mechanism of action for DAB/IL2.

We compared baseline FDG-PET and/or CT imaging to follow-up scans using qualitative radiology assessments which approximate Response Evaluation Criteria in Solid Tumors (RECIST) criteria [20]. Importantly, any new lesions that were identified would automatically signify progression. Figure 3A illustrates the following response rates: partial response (PR), 16.7% (10/60); stable disease (SD), 5% (3/60); mixed response (MR), 15% (9/60); and progressive disease (PD), 63.3% (38/60). We observed a marked improvement in the response rates of the 11 chemo/immuno-naïve patients: partial response (PR), 45.5% (5/11); stable disease (SD), 9.1% (1/11); mixed response (MR), 18.2% (2/11); and progressive disease (PD), 27.3% (3/11) (Figure 3A). Univariable and multivariable logistic modeling revealed a statistically significant increase in patients experiencing clinical benefit (PR+SD+MR) in the chemo/immuno-naïve population (p values = 0.011 and 0.017, respectively). We did not observe a decrease in response rate from prior exposure to IL-2 (p value = 0.389, univariable logistic model) which had been anticipated given the potential for cross-reactivity of antibodies between recombinant IL-2 and DAB/IL2 (data not shown). Stage IV melanoma is sub-classified into M1A (metastases to skin, subcutaneous tissues or distant lymph nodes; normal LDH), M1B (lung; normal LDH) and M1C (other visceral sites or distant metastases at any site; normal LDH; or any distant metastases; elevated LDH). We found that the partial response rate was highest in M1A patients (40%; Figure 3B and Table 3) and univariable logistic modeling indicated that the combined PR+SD+MR rate in the M1A population was higher than in the M1B population (p value = 0.028) and the M1C population (p value = 0.044). However, within the chemo/immuno-naïve population, the M1C patients experienced the greatest partial response rate (66.7%; Figure 3D). These data suggest that patients with the worst prognosis (i.e. M1C) seem to respond to DAB/IL2 at least as well as those with higher survival odds. No M1B patients had a partial or mixed response and only one did not progress (12.5%; Figure 3C). Last, although only two mucosal and two ocular melanoma patients were enrolled, we did observe 2/2 mixed responses and 1/2 mixed response in this small population, respectively (data not shown).

The median follow-up day from the first day of DAB/IL2 was 315 days (range from 28 to 1,198 days) for all patients and 995 days for seven patients who were alive at the time of the last follow-up (range from 631 to 1,198 days). The 1, 2, 3 and 4 year overall survival percentages were 40.0%, 17.9%, 9.2% and 9.2%, respectively. Although there appeared to be a trend towards improved overall survival in the chemo/immuno-naïve population (Figure 4A), the un-weighted log-rank test did not reveal a statistically significant difference (p value = 0.140). However, the overall survival probability was significantly higher in the patients in stage M1A compared to those in stage M1B (p value = 0.001), stage M1C (p value = 0.019) and combined stage M1B+M1C (p value = 0.010; Figure 4B) and the patients with a PR had a statistically significant longer overall survival time than those with the outcome PD (p value < 0.001; Figure 4C). Last, there appeared to be a trend towards decreased overall survival at year 2 in patients who had been previously administered recombinant IL-2 (data not shown), however this was not statistically significant (p value = 0.178).