Cancer Immunology, Immunotherapy

, Volume 58, Issue 8, pp 1297–1306 | Cite as

Therapeutic efficacy of ipilimumab, an anti-CTLA-4 monoclonal antibody, in patients with metastatic melanoma unresponsive to prior systemic treatments: clinical and immunological evidence from three patient cases

  • Anna Maria Di Giacomo
  • Riccardo Danielli
  • Massimo Guidoboni
  • Luana Calabrò
  • Dora Carlucci
  • Clelia Miracco
  • Luca Volterrani
  • Maria Antonietta Mazzei
  • Maurizio Biagioli
  • Maresa Altomonte
  • Michele Maio
Original Article

Abstract

The management of unresectable metastatic melanoma is a major clinical challenge because of the lack of reliably effective systemic therapies. Blocking cytotoxic T lymphocyte-associated antigen-4 (CTLA-4) has recently been proposed as a strategy to enhance cell-mediated immune responses to cancer, and clinical trials have demonstrated that anti-CTLA-4 therapy can produce durable outcomes with different response patterns than cytotoxic chemotherapy. We enrolled eight out of 155 patients with advanced melanoma in a multicentre phase II trial that evaluated the activity and tolerability of ipilimumab, a fully human, anti-CTLA-4 monoclonal antibody (www.clinicaltrials.gov; NCT00289627; CA184-008). Here we report our experience with three of these patients, who experienced progressive disease after a variety of previous therapies, including prior immunotherapies, and who achieved good outcomes with ipilimumab. One patient had a partial response ongoing at 17+ months on ipilimumab despite failure with four prior therapies, and the other two patients showed durable stable disease, both still ongoing at 17+ and 20+ months, respectively. The patient achieving a partial response experienced no side effects while receiving ipilimumab. The other two patients developed immune-related adverse events (irAEs) including rash (one case; grade 2) and diarrhoea (both cases; grades 1 and 2, respectively); the histopathology of colon biopsy samples from both was suggestive of colitis, with an abundant CD8+ T-cell infiltrate. Nausea, vomiting and acute pancreatitis were also observed in one patient. In addition, immunohistochemical findings of a dense CD8+, TIA1+ and granzyme B+ lymphoid infiltrate within a biopsied lesion provide indirect evidence of functional T-cell activation induced by treatment. These case reports highlight the potential for anti-CTLA-4-based therapy in previously treated patients with advanced melanoma. Moreover, because the patterns of response to ipilimumab differ from chemotherapy, we need to understand how and when patients may respond to treatment so that appropriate clinical decisions can be made.

Keywords

Ipilimumab Cytotoxic T-lymphocyte antigen-4 (CTLA-4) Metastatic melanoma Stable disease Immunotherapy Partial response 

Introduction

Unresectable metastatic melanoma has a poor prognosis, and its management is a major clinical challenge. The median survival for stage IV melanoma is typically 7–9 months, and 5-year survival is less than 10% [14]. Chemotherapy does not generally produce durable responses and has no beneficial effect on overall survival [14]. In contrast, one study showed that the majority of patients with metastatic melanoma who responded to high-dose bolus interleukin-2 (IL-2; 5/26 patients) had complete responses (4/5 patients), with 2 of the patients experiencing durable CR and all 4 CRs having stage M1a disease [18]. However, immunotherapy with IL-2 often results in substantial toxicity [2]. IL-2 must be administered in a hospital setting with an available intensive care facility and specialists skilled in cardiopulmonary or intensive care medicine, and it should only be offered to patients with a good performance status and in generally good health. While combining cytokines with chemotherapeutic agents achieves higher response rates compared with chemotherapy alone, toxicity is worsened and there is no improvement in overall survival [17]. Dacarbazine (DTIC) is still regarded as the standard choice for first-line treatment of most patients with metastatic melanoma, despite its lack of superiority compared with other treatments or best supportive care in phase III trials. In addition, no agent has been specifically approved for second-line therapy [1, 14]. It is clear that new treatment strategies which provide clinically meaningful benefits in first-line, as well as subsequent lines of therapy, are urgently required.

A recently proposed antitumour strategy involves augmentation of the cell-mediated immune system by blocking inhibitory pathways of T-cell activation [15]. The discovery that cytotoxic T-lymphocyte antigen-4 (CTLA-4) functions as a negative regulator of activated T-cells led to the hypothesis that blocking this molecule could potentiate T-cell activation and enable patients to mount a more effective antitumour immune response. The fully human, anti-CTLA-4 monoclonal antibody ipilimumab (Bristol-Myers Squibb, New York, NY, USA and Medarex, Inc., Princeton, NJ, USA) has been evaluated in phase I and II trials in patients with metastatic melanoma, and durable objective responses (ORs) have been observed [3, 8, 9, 10, 13, 19, 20]. These findings are particularly encouraging because they were often achieved in patients who were extensively pretreated and had evidence of progressive disease (PD) on or after prior therapy [11, 13, 22]. It is also apparent that in contrast to cytotoxic chemotherapeutic regimens where radiologic responses are typically observed soon after the start of treatment, clinical and radiologic response patterns with anti-CTLA-4 therapy can occur soon after therapy initiation or after several weeks or months in some patients [9, 11, 20, 22]. Moreover, our experiences as well as those of others from clinical trials with ipilimumab, indicate there are four broad categories of clinical and radiologic response: response in baseline lesions (captured by mWHO); stable disease with slow, steady decline in total tumor burden; response after initial increase in total tumor burden; and response in index and new lesions after the appearance of new lesions [9, 11, 16, 20].

Ipilimumab appears to be generally well tolerated [3, 8, 13, 21]. The adverse event profile, however, is different from chemotherapy, and most events are traced to the immune-related mechanism of action of these agents (immune-related adverse events [irAEs]). The irAEs reported with ipilimumab most frequently affect the skin, gastrointestinal tract, liver, and endocrine glands; grade 3/4 irAEs are mostly gastrointestinal (diarrhoea and colitis), and infrequent deaths have been reported (<1% of all cases). IrAEs are reversible in the majority of patients and manageable with treatment withdrawal or the use of corticosteroids/treatment guidelines; early treatment with high-dose steroids, which does not appear to affect the activity of ipilimumab, is critical to prevent the development of serious complications from irAEs [4, 8, 13].

This report presents the case records of three patients with progressive metastatic melanoma who, after being treated with one or more prior systemic therapies, were enrolled in a recently completed phase II study (http://www.clinicaltrials.gov; NCT00289627; CA184-008) [6]. One patient (Case 2) achieved a partial response (PR; first detected at week 12) and two (Cases 1 and 3) had durable SD after receiving induction therapy with ipilimumab scheduled at 10 mg/kg once every 3 weeks for four doses. Two patients (Cases 1 and 2) went on to receive maintenance therapy with ipilimumab (10 mg/kg every 12 weeks from week 24). The appearance of grade 3 or 4 toxicities precluded the third patient (Case 3) from entering the maintenance dosing phase; however, durable SD was still maintained without further dosing, 17+ months after the last dose of ipilimumab. While two of the patients described herein (Cases 1 and 3) experienced irAEs including rash (1 patient; grade 2) and diarrhoea (both patients; grades 1 and 2, respectively), the third patient (Case 2) did not experience any irAEs.

Case 1

Summary

This case illustrates ipilimumab-induced, durable SD. Prolonged SD with this novel class of antineoplastic therapies appears to be clinically meaningful, as patients with SD can survive months to years [9, 10, 20]. SD in patients treated with ipilimumab appears to be very different from what is observed in patients who have received chemotherapy. Although there is anecdotal evidence that SD may slowly evolve into an objective response (PR or even a CR) in patients who received ipilimumab [9], this patient experienced SD for greater than 20 months.

In May 2005, a 76-year-old Caucasian female had a pigmented skin lesion excised from her left leg. Pathologic assessment revealed a pT3N1aMx melanoma; there was superficial spread and a vertical growth pattern (Clark Level IV), regression and invasion of vascular spaces, tumour-free margins and tumour-infiltrating lymphocytes in a non-brisk pattern. In the same month, a sentinel-node biopsy of the left inguinal node revealed melanoma micrometastases, while the subsequent left inguinocrural lymph-node dissection did not show any additional metastatic nodes.

In October 2005, multiple pigmented lesions, ranging from 3 to 10 mm in diameter, around the excision site on the third distal portion of the left leg were investigated and pathologic assessment confirmed metastasis of melanoma. In December 2005, the patient was enrolled in a phase II, randomised clinical trial; she received six cycles of chemoimmunotherapy comprising DTIC 800 mg/m2 i.v. on Day 1, the immunotherapies interferon-α (IFN-α) 3 MU s.c. on Days 11 and 18 and thymosin-α1 6.4 mg s.c. on Days 8–11 and 15–18. In June 2006, cutaneous PD was confirmed. From June until August 2006, she received DTIC 800 mg/m2 i.v. on Day 1 of a 28-day cycle for three cycles. Further cutaneous PD was confirmed in August 2006.

In September 2006, she was enrolled into another clinical trial (Protocol CA184-008) and began induction therapy with ipilimumab 10 mg/kg, i.v., once every 3 weeks for four doses (weeks 1, 4, 7 and 10). At enrollment, her Eastern Cooperative Oncology Group performance status (ECOG PS) was 0. The first tumour assessment at week 12 showed SD and this was still ongoing at week 24. The cutaneous leg lesions showing a sessile base before treatment, progressively merged into a single, larger lesion during therapy with the tendency to become semi-pedunculated (Fig. 1).
Fig. 1

Case 1 Cutaneous leg lesions showing a pretreatment sessile base (b) evolved into a single lesion during therapy with ipilimumab (week 12 of trial CA184-008; c with a progressive tendency to become semi-pedunculated (week 12 of trial CA184-025 and 14 months after starting treatment; d During treatment, there was re-epithelialisation of the lesion stalk and a dense mononuclear infiltrate confined to the body of the tumour (a; H&E; ×25). Four months after surgical excision, the lesion’s base completely re-epithelialised showing no macroscopic evidence of disease (week 36 of trial CA184-025 and 19 months after starting treatment; e)

In March 2007 (week 24 of the trial), she entered the maintenance phase of the trial and received her first maintenance dose of ipilimumab 10 mg/kg every 12 weeks. The closure of trial CA184-008 meant she was transferred to a roll-over trial (http://www.clinicaltrials.gov; NCT00162123; CA184-025) to continue maintenance therapy [7]. In June 2007, the patient received her second maintenance dose of ipilimumab, and her CT scan showed ongoing SD, 11 months after starting treatment. The leg neoplastic lesions shown in Fig. 1 showed a slight enlargement during the early phases of treatment, merging into a single lesion; however, after this initial progression, the lesion remained stable during the next 6 months. In October 2007, after 12 weeks in trial CA184-025 and 15 months after starting ipilimumab treatment, the lesion was excised because of bleeding subsequent to mechanical trauma. Histopathology showed a dense inflammatory cell infiltrate containing a high proportion of activated cytotoxic T lymphocytes (Fig. 2).
Fig. 2

Case 1 Cutaneous leg lesion excised during the maintenance phase of ipilimumab therapy in the CA184-025 roll-over trial, illustrating a dense inflammatory infiltrate (a) containing a high proportion of activated cytotoxic lymphocytes tending to dissociate individual melanoma cells (bd: CD8, TIA-1 and granzyme-B immunostainings, respectively; ×400). Small haemorrhages were also detected within the lesion (indicated by the arrow in a; H&E; ×100)

The patient tolerated ipilimumab treatment well, reporting only two AEs that were considered treatment-related. At week 7, she experienced grade 1 diarrhoea that resolved on the same day without treatment. At week 10, she developed a grade 2 skin rash on her left thigh. The rash resolved within 3 weeks following treatment with oral corticosteroids and antihistamines (oral prednisone 5 mg three times daily and clorfenamine 4 mg once daily). The corticosteroid treatment was continued and then tapered, stopping at week 28 with no recurrence of the rash. In January 2007, week 18 of treatment, the patient underwent a colonoscopy (which was not part of routine care). It was performed at week 18 since the patient was experiencing softened stools without an increase in the frequency over baseline (that could classify her as grade 1 diarrhoea). The colon biopsy samples analysed were reminiscent of ulcerative colitis, with moderate to severe inflammatory infiltrates confined to the mucosa and submucosa (Fig. 3). Treatment was not considered necessary.
Fig. 3

Case 1 Colon biopsies taken after completing induction therapy with ipilimumab (week 18). Histopathology was reminiscent of ulcerative colitis with moderate to severe inflammatory infiltrates confined exclusively to the mucosa and submucosa (a; H&E; ×100). Higher magnification demonstrated neutrophils infiltrating crypts and forming crypt microabscesses (b; H&E; ×400). Consistent levels of intraepithelial CD8+ T-lymphocytes were evident (c; CD8 immunostaining; ×400)

The patient’s SD is still ongoing, 20+ months after starting ipilimumab treatment.

Case 2

Summary

This patient achieved an ongoing PR and had no irAEs. This is an interesting case given the controversy surrounding the hypothesis that the appearance of irAEs correlates with anti-CTLA-4 activity. Because the mechanism of action of anti-CTLA-4 monoclonal antibodies is continued blockade of immune suppression, it is easy to conceive that CTLA-4 blockade-mediated irAEs may be linked with antitumor clinical activity. Published data suggest a correlation between irAEs and ipilimumab activity [3, 4, 21]. However, some patients achieve a response or SD without an irAE, as is demonstrated in this case. Therefore, predicting irAEs may not be as straightforward as previously thought and further investigation into the potential correlation between ipilimumab activity and irAEs is warranted.

In January 2004, a 42-year-old Caucasian male presented with growth of a pigmented skin lesion on the left leg and a homolateral inguinal lymph node. The lesion was resected in February of the same year and pathologic analysis showed a skin melanoma with superficial spreading and ulceration, Breslow thickness 7.00 mm, Clark level V and no tumour regression or venous invasion. Left inguinal lymphadenectomy was performed and metastatic melanoma was detected in one of the nine nodes that were removed and examined; the final pathologic staging was pT4bN1bMx.

From March 2004 until November 2005 he received IFN-α (2.5–5 MU twice weekly). In August 2005, two subcutaneous nodules were removed from his left thigh and pathologic analysis confirmed metastasis of melanoma. In November 2005, lower left leg hyperthermic perfusion with melphalan was performed because of subcutaneous PD. However, as a consequence of this, the patient had rhabdomyolysis of the left thigh leading to acute renal failure that required dialysis. In February 2006, a whole-body CT scan showed no evidence of metastasis. However, at the end of March, multiple subcutaneous relapses of disease in the lower left leg were noted during a positron emission tomography (PET) scan.

In May 2006, the patient underwent vaccination with MAGE-3 protein every 2 weeks for six doses. A whole-body CT scan in July showed ‘slow’ PD; thus, he remained on treatment and received a second cycle of immunotherapy with MAGE-3 protein every 3 weeks.

In September 2006, another CT scan showed PD on the left thigh and from October he received three cycles of chemoimmunotherapy comprising DTIC 800 mg/m2 i.v. on Day 1 and thymosin-α1 3.2 mg s.c. on Days 8–11 and 15–18 every 4 weeks. In December 2006, a CT scan showed PD in the left leg, lymph nodes, liver and spleen and a PET scan showed pathologic lesions in the left thigh, and in the right inguinal, iliac, and lumboaortic lymph nodes.

In January 2007, the patient was enrolled in the CA184-008 ipilimumab clinical trial and he started induction therapy (ipilimumab 10 mg/kg, i.v., once every 3 weeks for four doses—weeks 1, 4, 7 and 10). The patient had an ECOG PS of 1 at enrollment. After all four doses of ipilimumab, the tumour assessment at week 12 showed a PR relative to the baseline tumor assessment (prior to treatment). There was a 36% increase in tumor volume from weeks 12 to 16, which by standard WHO criteria is considered PD. However, per CA184-008 study protocol, the patient was followed beyond PD with a confirmatory scan 4 weeks later. A response of PR was observed again at weeks 20 and 24 relative to baseline. Noteworthy, tumour assessment at week 12 showed shrinkage of index liver metastases (Fig. 4c) together with an increase of non-index spleen metastases (Fig. 4d). PET and CT scans also demonstrated the disappearance of many subcutaneous lesions on the left thigh after ipilimumab induction therapy (Fig. 5).
Fig. 4

Case 2 CT scans illustrating the disease status before (a, b) treatment with ipilimumab, at week 12 from the beginning of treatment (c, d), and at week 24 of the CA184-025 roll-over trial (e, f). After the induction phase, the patient experienced substantial shrinkage of liver metastases at segments II and VI (c), while two pre-existing splenic lesions progressed (d). At week 24 of the roll-over trial, he experienced further shrinkage of hepatic lesions (e), disappearance of one splenic nodule and significant shrinkage of the second one (f)

Fig. 5

Case 2 CT and PET scans illustrating the status of subcutaneous metastases before (a, c) and after treatment (week 20; b, d) with ipilimumab. The arrows on the pretreatment scans show the sites of metastases. The patient achieved a PR, with the disappearance of many subcutaneous lesions on the left thigh

In July 2007, he received his first maintenance dose of ipilimumab (10 mg/kg) in the CA184-025 roll-over trial and was scheduled to receive further 10 mg/kg doses every 12 weeks. In October 2007, the patient remained in PR, 9 months after starting ipilimumab. Additionally, tumour assessment performed at week 24 after his first maintenance dose, showed further reduction of hepatic lesions (Fig. 4e), together with shrinkage of previously progressing non-index spleen lesions (Fig. 4f). He has a good PS (ECOG PS 1), has not suffered any irAEs while receiving ipilimumab and has a PR ongoing 17+ months after starting ipilimumab.

Case 3

Summary

This case illustrates the management of a patient with prolonged SD and ipilimumab-related pancreatitis and biliary toxicities. Although diarrhoea is a frequent symptom associated with ipilimumab therapy, common symptoms such as abdominal pain may also indicate more serious complications such as immune-related pancreatitis. It is important to ensure that the patient is made aware of all potential irAE symptoms and encouraged to report his/her symptoms early, in order to receive immediate treatment and prevent potentially serious complications. Laboratory tests to detect changes in the levels of pancreatic, hepatobiliary and liver enzymes should be run when warranted. If the patient’s enzyme levels are sufficiently elevated and the patient is symptomatic, ipilimumab therapy should be interrupted. Patients should be hospitalised and treated with parenteral and/or oral corticosteroids as medically indicated to manage inflammation.

In April 2000, a 36-year-old Caucasian female who had experienced acute gallstones-related pancreatitis the previous year had a pigmented lesion excised from her left ankle. Pathologic assessment revealed a cutaneous melanoma with superficial spreading, Clark level IV and thickness 2.6 mm, with a negative sentinel-node biopsy (pT3N0Mx); she, therefore, received adjuvant therapy with IFN-α (3 MU three times per week) for 18 months. In February 2002, a histologically confirmed amelanotic relapse appeared in the region of the primary site during therapy, followed by multiple relapses in the same location that were surgically removed.

In August 2004, a whole-body PET scan revealed areas with moderate hypermetabolism in the third distal portion of the right leg. Three months later, the right obturator lymph nodes (1 N+) were excised, and she received lower limb hyperthermic perfusion with melphalan. The left inguinal lymph nodes (1 N+) were excised in January 2005, and neuropathy and lymphedema of the right leg were subsequently observed. In June 2006, subcutaneous progression in the lower right leg was observed. In October 2006, she started chemoimmunotherapy with DTIC 800 mg/m2 i.v. on Day 1 and thymosin-α1 3.2 mg s.c. on Days 8–11 and 15–18 every 28 days for three cycles; but in December 2006, PD was observed in the lower right leg and treatment was stopped.

In January 2007, she was enrolled in trial CA184-008 to receive induction therapy with ipilimumab 10 mg/kg i.v. once every 3 weeks for four doses (weeks 1, 4, 7 and 10). Grade 2 nausea and grade 2 vomiting were recorded on three consecutive days between the second and third doses; both resolved with symptomatic therapy. No elevation in liver function tests and pancreatic enzymes were observed between the second and third dose. In February 2007 (week 7), the patient was asymptomatic for any irAEs and was therefore infused with the third dose of ipilimumab; later that day she reported abdominal pain. Laboratory test results from her week-7 visit subsequently revealed grade 3 and 4 increases in amylase and lipase, respectively. Because of her past medically history of acute pancreatitis and elevated enzymes in the presence of clinical symptoms, the patient was suspected of having immune-related pancreatitis and ipilimumab treatment was interrupted at her week-10 visit (March 2007); She continued, however, to be followed up for tumour assessment. She started therapy with intramuscular dexamethasone 8 mg that was later switched to oral prednisone 25 mg and gradually tapered over a month. Blood tests taken several days later were positive for anti-pancreas auto-antibodies (islet cell antibodies and glutamic acid decarboxylase antibodies). In March 2007 (week 12), she reported grade 2 diarrhoea (the patient refused colonoscopy), and a whole-body CT scan showed no visceral evidence of disease (despite cutaneous lesions) and the patient was classified as having SD; there was no significant change in the gastrointestinal tract.

In late March 2007 (week 13) she was hospitalised because of persistent abdominal symptoms. Ultrasound and X-ray scans excluded pancreatitis, although biliary sand was present in the gallbladder, and levels of transaminases, amylase and lipase were moderately elevated. She was discharged in good health 4 days later, although abdominal symptoms recurred and magnetic resonance cholangiopancreatography, performed in late April 2007, showed evidence of recent acute pancreatitis involving the body and tail of the pancreas. In early May 2007 (week 16), clinical examination showed cutaneous SD and CT scan was still negative; diarrhoea and abdominal pain were absent.

The grade 2 diarrhea reported in March 2007 indicated that the patient might have developed an immune-related intestinal disorder. In May 2007 (week 16), a colonoscopy was performed (in the absence of significant clinical symptoms) which showed no macroscopic alterations, but histologic examination of colonic biopsy showed features suggesting immune-related colitis (not shown). Amylase and lipase levels were stable and within the normal physiologic range. In late May, the patient stopped oral prednisone because she was no longer symptomatic.

In June 2007 (week 20), the patient had ongoing SD relative to the baseline tumor assessment (prior to treatment). At week 24, there was evidence of PD in the inguinal node. In September 2007 (week 32) a cutaneous thigh lesion was resected and immunohistochemical staining showed melanoma-cell nests intermingled with an abundant mononuclear cell infiltrate, comprising a relatively high number of activated cytotoxic lymphocytes (Fig. 6). In late October 2007 (week 40), there was no evidence of PD since the inguinal node detected at week 24 was not confirmed by CT scan. The patient remains in SD, 17+ months after starting ipilimumab.
Fig. 6

Case 3 Cutaneous thigh lesion resected at week 32. Melanoma-cell nests are intermingled with an abundant mononuclear cell infiltrate (a, H&E; ×100), comprising a relatively high number of activated cytotoxic lymphocytes, as shown by CD8 (b), the apoptosis promoting protein TIA-1 (c) and granzyme B (d) immunohistochemical staining; ×400

Discussion

Standard first-line systemic therapy for metastatic melanoma fails to provide meaningful or durable clinical benefit in the vast majority of cases and, as a consequence, there are substantial numbers of patients eligible for second-line treatment. No agent has been specifically approved in this setting [1].The effects of ipilimumab have been investigated in a number of clinical trials involving previously treated and chemotherapy-naïve patients, and durable responses and SD have been achieved [3, 8, 9, 10, 13, 19, 20]. Our three cases confirm that durable disease control (PR or SD) can be achieved with ipilimumab in patients showing progression on numerous prior systemic treatments including prior immunotherapies. Case 2, for instance, received long-term immunotherapy with IFN-α, chemotherapy with melphalan, chemoimmunotherapy with DTIC plus thymosin-α1 and vaccination with MAGE-3 protein in a clinical trial, all without clinical benefit. While caution must be exercised in extrapolating from three case reports, our observations are in-line with prior reports and suggest that anti-CTLA-4 therapy, with different patterns of response than conventional chemotherapy, may provide a valid therapeutic option in patients with melanoma that is refractory to a variety of other treatments.

A characteristic finding in trials of ipilimumab in metastatic melanoma is the durability of responses and SD observed in certain subsets of patients. Recent pooled analysis of six clinical trials, for example, showed that ipilimumab given alone or with other agents produced ORs ranging from 3 months to 4+ years in 45/356 patients (12.6%) [9]. Eighteen of the 45 patients (40%) had SD ranging from 2 to 16 months prior to achieving their OR, and 84 additional patients experienced SD as their best overall response. SD was durable (≥24 weeks) in 23/84 patients (27%) and ongoing in 33 patients at the time of analysis [9]. Disease control (PR or SD) in our patients who have failed several prior treatments is ongoing from 17+ and 20+ months after starting ipilimumab. Durable responses in a second-line setting have also been recently observed in a phase II study of IL-2 immunotherapy in 26 patients with metastatic melanoma, with two of five patients having achieved an objective response lasting 26+ and 41+ months [18]. Taken together, it seems likely that the immune-related mechanisms of action of these approaches may contribute to the durable outcomes. Furthermore, by using ipilimumab as maintenance therapy, it is hoped that the immune system can be periodically ‘boosted’ to maintain antitumour activity, thereby resulting in prolonged disease stabilisation.

Evidence of the immunopotentiating mechanism of action of ipilimumab, achieved through CTLA-4 inhibition, can be seen in the hematoxylin and eosin (H&E) and immunohistochemical staining of tumour sections from Cases 1 (Fig. 2) and 3 (Fig. 6). In both tumours there is a dense immune-cell infiltrate that contains a high proportion of cytotoxic T-cells. Combined with the presence of the apoptosis promoting protein TIA-1 and granzyme B, the protein which is crucial for the rapid induction of target-cell apoptosis by cytotoxic T-cells, these results suggest that immune-mediated tumour cell death has occurred in these patients.

The best overall response in two of our three patients was SD (ongoing at 17+ and 20+ months), and similar disease control has been seen in patients treated with ipilimumab in previous studies; a pooled analysis of clinical trial data has shown that 23/356 patients (6.3%) had durable SD (≥24 weeks) [9]. In contrast to the transient SD that may occur in cancer patients treated with cytotoxic chemotherapy, the durable SD associated with anti-CTLA-4 therapy should be considered a meaningful outcome in patients with refractory metastatic melanoma who have a very poor prognosis. Furthermore, there is some evidence that patients receiving anti-CTLA-4 therapy can have prolonged periods of SD, or even apparent PD, prior to developing an OR [9, 11, 16, 19, 22]. Data from the phase II/III trial programme will help to determine the overall importance of durable SD in metastatic melanoma, and to investigate whether maintenance therapy may prolong disease containment or eventually induce an OR in patients who may not achieve PR/CR by week 12 when the induction phase is complete.

The heterogeneity of response patterns seen with ipilimumab may reflect differences in immune systems among patients that may affect the time taken for CTLA-4 blockade to generate a sustainable antitumour immune response. In our cases, PR or SD was observed at the first assessment scan following induction therapy, although in Case 2 there was fluctuation in the level of disease control, with PR observed following the first assessment scan at week 12, increased tumor burden from week 12 to 16, and a return to PR at week 20 and thereafter. In Case 3, there was unconfirmed disease progression at week 24, although several months later, the patient remains in ongoing SD (17+ months). Unlike chemotherapy, anecdotal evidence suggests that the emergence of new lesions during ipilimumab treatment does not necessarily indicate treatment failure [22]. In our experience, PD can occur in non-target lesions, while initial target lesions continue to regress or remain stable. Subsequently, lesions that have previously progressed may regress (Fig. 4); as it will take time for the immune system to recognise and control melanoma lesions. Clinically this would manifest as PD prior to becoming a PR or SD. Based on these observations it appears appropriate that substantial PD, which may not be reflected in a patient’s performance status and/or laboratory tests, is confirmed before considering a new therapy. Further clinical data will help define when alternative therapies should be considered.

IrAEs have been well characterised with ipilimumab and are generally mild to moderate; grade 3/4 irAEs are mostly gastrointestinal (diarrhoea and colitis), and infrequent deaths have been reported (<1% of all cases). It is critical to treat irAEs promptly with high-dose steroids because events such as diarrhoea, endocrinopathies, and immune hepatitis have the potential to quickly develop into serious, potentially life-threatening AEs if not diagnosed and treated in a timely manner. Prophylactic steroid use may reduce the incidence of Grade 3/4 irAEs, but does not appear to affect the antitumor activity of ipilimumab [12]. The irAEs observed in our study patients are consistent with the profile defined in clinical trials [3, 4, 8, 16, 19]. One patient had no AEs during ipilimumab treatment. Two patients reported diarrhoea, a frequently reported irAE, and colonic histopathology was reminiscent of immune-related colitis. One of these patients also developed a skin rash and the other experienced nausea, vomiting and pancreatitis. The relationship between administration of ipilimumab and development of pancreatitis in Case 3 was unclear, particularly in view of her previous clinical history, but it was nevertheless considered prudent that she should not continue with ipilimumab therapy beyond the induction phase.

Both patients with irAEs were successfully treated with corticosteroids. Preclinical studies indicate corticosteroids do not affect the efficacy of anti-CTLA-4-mediated tumour inhibition [5] and a similar outcome has been reported in a clinical study [4]. Two of three patients presented in this report were given steroids for the management of their irAEs and it did not appear that steroids interfered with the clinical benefit the patients experienced with ipilimumab. Against this background, steroids are recommended for the management of moderate-severe irAEs [13, 15]. In our experience, most irAEs likely to occur with ipilimumab can be resolved successfully providing they are identified and managed promptly and appropriately.

One must be cautious in extrapolating from three case reports, but our observations demonstrate that induction therapy with 10 mg/kg ipilimumab is well tolerated and produces durable disease control (OR or SD) in patients with metastatic melanoma unresponsive to a variety of prior therapeutic modalities. The optimal use of these agents requires clinicians to have a full appreciation of their patterns of response and tolerability profile, to understand how and why they differ from other therapies, including chemotherapy, and how these factors will affect the making of clinical decisions. We eagerly anticipate the results of ongoing clinical trials with ipilimumab.

Notes

Acknowledgments

Medical writing and editorial assistance provided by Gardiner-Caldwell US, with financial support from Bristol-Myers Squibb. This work was supported in part by grants awarded to MM from the Associazione Italiana per la Ricerca sul Cancro, and Istituto Superiore di Sanità—Alleanza Contro il Cancro for the project “Rete nazionale per studi clinici e di strutture GMP per le bioterapie dei tumori”. The authors wish to acknowledge the excellent nursing support of Angela Iacovelli, Marilena Piccinelli, Massimo Resti, and Sergio Speranza.

References

  1. 1.
    Agarwala SS (2004) Relevance and necessity of studies on second-line chemotherapy in melanoma. Onkologie 27:527–528PubMedCrossRefGoogle Scholar
  2. 2.
    Atkins MB, Lotze MT, Dutcher JP, Fisher RI, Weiss G, Margolin K, Abrams J, Sznol M, Parkinson D, Hawkins M, Paradise C, Kunkel L, Rosenberg SA (1999) High-dose recombinant interleukin 2 therapy for patients with metastatic melanoma: analysis of 270 patients treated between 1985 and 1993. J Clin Oncol 17:2105–2116PubMedGoogle Scholar
  3. 3.
    Attia P, Phan GQ, Maker AV, Robinson MR, Quezado MM, Yang JC, Sherry RM, Topalian SL, Kammula US, Royal RE, Restifo NP, Haworth LR, Levy C, Mauroukakis SA, Nichol G, Yellin MJ, Rosenberg SA (2005) Autoimmunity correlates with tumor regression in patients with metastatic melanoma treated with anti-cytotoxic T-lymphocyte antigen-4. J Clin Oncol 23:6043–6053PubMedCrossRefGoogle Scholar
  4. 4.
    Beck KE, Blansfield JA, Tran KQ, Feldan AL, Hughes MS, Royal RE, Kammula US, Topalian SL, Sherry RM, Kleiner D, Quezado M, Lowy I, Yellin M, Rosenberg SA, Yang JC (2006) Enterocolitis in patients with cancer after antibody blockade of cytotoxic T-lymphocyte-associated antigen 4. J Clin Oncol 24:2283–2289PubMedCrossRefGoogle Scholar
  5. 5.
    Chen B, Phillips J, Greenbaum M, Klitzing D, Cardarelli PM, Korman A (2007) Efficacy of anti-CTLA-4 antibody in the Sa1 N tumor model when combined with dexamethasone (abstract 2202). In: Presented at the AACR Annual Meeting, 14–28 April 2007, Los Angeles, CA, USAGoogle Scholar
  6. 6.
    ClinicalTrials.gov. A single arm study of ipilimumab monotherapy in patients with previously treated unresectable stage III or IV melanoma. http://clinicaltrials.gov/ct/show/NCT00289627
  7. 7.
    ClinicalTrials.gov. A companion study for patients enrolled in prior/parent ipilimumab studies. http://clinicaltrials.gov/show/NCT00162123
  8. 8.
    Fischkoff SA, Hersh E, Weber J, Powderly J, Khan K, Pavlick A, Samlowski W, O’Day S, Nichol G, Yellin M (2005) Durable responses and long-term progression-free survival observed in a phase II study of MDX-010 alone or in combination with dacarbazine (DTIC) in metastatic melanoma (abstract 7525). J Clin Oncol 23(suppl)Google Scholar
  9. 9.
    Hamid O, Urba WJ, Yellin M, Nichol GM, Weber J, Hersh EM, Tchekmedyian S, Hodi FS, Weber R, O’Day S (2007) Ipilimumab (MDX-010) in patients with stage III/IV melanoma: kinetics and duration of response. Eur J Cancer Suppl 5:396CrossRefGoogle Scholar
  10. 10.
    Hersh EM, O’Day SJ, Powderly J, Khan KD, Pavlick AC, Cranmer L, Samlowski WE, Nichol GM, Yellin MJ, Fischkoff SA, Weber JS (2008) Disease control and long-term survival in chemotherapy-naive patients with advanced melanoma treated with ipilimumab with or without dacarbazine. J Clin Oncol (submitted)Google Scholar
  11. 11.
    Hodi FS, The Global Ipilimumab Melanoma Study Group (2008) Novel efficacy criteria for antitumor activity to immunotherapy using the example of ipilimumab, an anti-CTLA-4 monoclonal antibody (abstract 3008). J Clin Oncol 26(suppl)Google Scholar
  12. 12.
    Lin R, Yellin MJ, Lowy I, Safferman A, Chin K, Ibrahim R (2008) An analysis of the effectiveness of specific guidelines for the management of ipilimumab-mediated diarrhea/colitis: prevention of gastrointestinal perforation and/or colectomy (abstract 9063). J Clin Oncol 26(suppl)Google Scholar
  13. 13.
    Maker AV, Yang JC, Sherry RM, Topalian SL, Kammula US, Royal RE, Hughes M, Yellin MJ, Haworth LR, Levy C, Allen T, Mavroukakis SA, Attia P, Rosenberg SA (2006) Intrapatient dose escalation of anti-CTLA-4 antibody in patients with metastatic melanoma. J Immunother 29:455–463PubMedCrossRefGoogle Scholar
  14. 14.
    O’Day S, Boasberg P (2006) Management of metastatic melanoma 2005. Surg Oncol Clin N Am 15:419–437PubMedCrossRefGoogle Scholar
  15. 15.
    O’Day SJ, Hamid O, Urba WJ (2007) Targeting cytotoxic T-lymphocyte antigen-4 (CTLA-4): a novel strategy for the treatment of melanoma and other malignancies. Cancer 110:2614–2627PubMedCrossRefGoogle Scholar
  16. 16.
    Saenger Y, Wolchok J (2008) The heterogeneity of the kinetics of response to ipilimumab in metastatic melanoma: patient cases. Cancer Immun 8:1PubMedGoogle Scholar
  17. 17.
    Sasse AD, Sasse EC, Clark LGO, Ulloa L, Clark OAC (2007) Chemoimmunotherapy versus chemotherapy for metastatic malignant melanoma. Cochrane Database of Systematic Reviews 2007, Iss 1. Art No: CD005413. doi:10.1002/14651858.CD005413.pub2
  18. 18.
    Tarhini AA, Kirkwood JM, Gooding WE, Cai C, Agarwala SS (2007) Durable complete responses with high-dose bolus interleukin-2 in patients with metastatic melanoma who have experienced progression after biochemotherapy. J Clin Oncol 25:3802–3807PubMedCrossRefGoogle Scholar
  19. 19.
    Weber J (2007) Anti-CTLA-4 antibody ipilimumab: case studies of clinical response and immune-related adverse effects. Oncologist 12:864–872PubMedCrossRefGoogle Scholar
  20. 20.
    Weber JS, Hersh EM, Yellin M, Nichol GM, Urba W, Powderly JD, O’Day SJ (2007) The efficacy and safety of ipilimumab (MDX-010) in patients with unresectable stage III or stage IV malignant melanoma [abstract 8523]. J Clin Oncol 25(suppl)Google Scholar
  21. 21.
    Weber JS, Targan S, Scotland R, Snively J, Garcia M, Yellin M, Fischkoff S, Nichol G (2006) Phase II trial of extended dose anti-CTLA-4 antibody ipilimumab (formerly MDX-010) with a multi-peptide vaccine for resected stages IIIC and IV melanoma (abstract 2510). J Clin Oncol 18(suppl)Google Scholar
  22. 22.
    Wolchok JD, The Global Ipilimumab Melanoma Study Group (2008) Antitumor response and new lesions in advanced melanoma patients on ipilimumab treatment (abstract 3020). J Clin Oncol 26(suppl)Google Scholar

Copyright information

© Springer-Verlag 2008

Authors and Affiliations

  • Anna Maria Di Giacomo
    • 1
  • Riccardo Danielli
    • 1
  • Massimo Guidoboni
    • 1
  • Luana Calabrò
    • 1
  • Dora Carlucci
    • 1
  • Clelia Miracco
    • 2
  • Luca Volterrani
    • 3
  • Maria Antonietta Mazzei
    • 3
  • Maurizio Biagioli
    • 4
  • Maresa Altomonte
    • 1
  • Michele Maio
    • 1
    • 5
  1. 1.Division of Medical Oncology and Immunotherapy, Istituto Toscano TumoriUniversity Hospital of SienaSienaItaly
  2. 2.Department of Pathology, Istituto Toscano TumoriUniversity Hospital of SienaSienaItaly
  3. 3.Department of Radiology, Istituto Toscano TumoriUniversity Hospital of SienaSienaItaly
  4. 4.Division of Dermatology, Istitutto Toscano TumoriUniversity Hospital of SienaSienaItaly
  5. 5.Cancer Bioimmunotherapy Unit, Department of Medical Oncology, Centro di Riferimento OncologicoIstituto di Ricovero e Cura a Carattere ScientificoAvianoItaly

Personalised recommendations