Abstract
Two papers, each with an accompanying editorial, published recently in the New England Journal of Medicine, shattered the stagnant waters of melanoma therapeutics. 1–4
The first 1,2 made a bold claim that survival of patients with metastatic melanoma could be improved with treatment with Ipilimumab, a monoclonal antibody that potentiates anti-tumour immunity, despite only a small number of treated patients experiencing a significant resolution of their tumours.
The second paper 3,4 claimed instead an audacious 81% response rate resulting in major regression of large tumour volumes achieved with a new drug, Plexxikon (PLX4032 or RG7204), that targets the V600E B-RAF mutation harboured by some patients with metastatic melanoma. Despite such major responses an impact on survival remains at this point uncertain.
These observations for a cancer perceived by many to be refractory to all treatment merit a critical appraisal.
Ipilimumab
Patients in the Ipilimumab trial had inoperable stage III (unresectable loco-regional recurrences) or stage IV (metastases beyond regional lymph nodes) melanoma, progressing against previous treatment. The study compared, in 676 selected HLA-A*0201-positive patients, Ipilimumab, with a vaccine, glycoprotein 100 (gp100) (Box 1) administered in a randomized fashion on its own, or in combination with Ipilimumab.
Molecular aspects of the studies under discussion
Ipilimumab targets the cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4). This molecule, the CTLA-4, operates as an immune checkpoint that down regulates pathways of T-cell activation and prevents autoimmunity. If this function is blocked – as is the case with Ipilimumab – the anti-tumour T-cell response is potentiated. 1,2 At the same time the risk of autoimmune events increases.
gp100 are antigenic peptides recognized by cytotoxic T lymphocytes, that belong to the class of melanocyte differentiation antigens, present in both melanoma cells and normal melanocytes. They are restricted to the HLA-A*0201. Because of the comparator gp100 arm in the Ipilimumab study, all 676 patients enrolled had to have a positive status for HLA-A*0201. 1,2 It is estimated that the HLA-A*0201 allele is expressed in approximately 50% of Caucasian populations. 5
B-RAF is the most frequently mutated protein kinase in human cancers. 3 Approximately 60% of melanomas carry B-RAF mutations. Ninety percent of reported B-RAF mutations in this tumour result in a substitution of glutamic acid for valine at amino acid 600, known as the V600E mutation. B-RAF-mutant melanomas are highly dependent on B-RAF kinase activity that drives the tumour to proliferation. 3 Plexxikon (PLX4032 or RG7204) blocks this activity.
The original primary end-point of this phase III trial was the best overall objective response rate – complete or partial shrinkage of tumour masses. During the course of the trial the investigators found it necessary to amend this primary end-point, to overall survival instead. 1
The primary comparison in overall survival was between the Ipilimumab-plus-gp100 group and the gp100-alone group. Comparison with conventional chemotherapy or an observation arm was not part of the design of this trial.
The authors reported that patients who received Ipilimumab with or without the gp100 vaccine survived nearly 4 months longer than did those who received the gp100 vaccine alone with a median survival from randomization, of 10 months and 6.4 months, respectively. There were three (1.7%) complete responses among patients receiving Ipilimumab and none among patients receiving gp100 alone. 1
Considering the molecular target of Ipilimumab (Box 1), treatment was associated with adverse immune-related events, some severe, but most were reportedly reversible with appropriate treatment. Nevertheless, there were 14 deaths related to the study drugs with a mortality of 2.1%. 1
Inhibition of mutated, activated B-RAF with Plexxikon (PLX4032 or RG7204)
PLX4032, a new drug taken by mouth that inhibits the activity of the V600E B-RAF mutation (Box 1) was evaluated in 32 patients with metastatic melanoma harbouring the V600E mutation. Twenty-six of these patients (81%) experienced a complete (2) or partial shrinkage (24) of their tumours. 3
This remarkable result, for a cancer commonly perceived as refractory to treatment, generated understandable excitement in medical circles and wide publicity in the media.
A dermatologist reviewing the study recently at an international conference described the results as ‘totally incredible in the field of melanoma’ 6 and a BBC commentator asked if this was the ‘penicillin moment for cancer’. 7
Representative responses illustrated in the published article from this extended phase I study, are indeed spectacular and were seen at all sites of metastatic disease including the liver, small bowel and bone. 3 They lasted for an average of 7 months before disease progressed again. 3
Side-effects associated with Plexxikon included rash, arthralgia, fatigue and hyperkeratotic skin lesions including frank low-grade squamous carcinoma of the skin in 31% of patients. 3 At this point it is not known if squamous carcinoma can develop in organs other than the skin. However the drug was generally well-tolerated and its oral administration is a distinct advantage.
Both studies are at the centre of an ongoing debate on research methodologies for the development of new treatments for cancer and raise important clinical and ethical issues. 8–12
Should priority be given to a treatment that appears to prolong survival despite the absence of substantial objective tumour responses or should the focus be on a drug that achieves meaningful resolution of metastases in the majority of treated patients but, to date, with no demonstrable impact on survival?
Will these drugs with a different mode of action achieve anti-tumour synergy if combined in the future or will their toxicities prohibit such combination? Should they be administered in sequence and in what order?
Arguably tumour response in phase II trials is one of the best predictors of a drug's activity and remains to date a highly effective and valid criterion for the development of anti-cancer treatments. 10
This same criterion, the best overall response rate, was the primary end point in the Ipilimumab study until the investigators amended it to overall survival. 1
‘Overall survival’ in the Ipilimumab trial is defined as the time from randomization to death from any cause. 1 It is not overall survival from initial diagnosis of melanoma or indeed from the development of stage III or IV disease – important landmarks in the evolution of this cancer and the last two, criteria for enrolment into the study. The disease-free interval before the development of clinically apparent regional or distant metastases may influence survival from these stages of melanoma. 13–15 It would therefore be reassuring to know that the survival differences that favour Ipilimumab were not influenced by imbalances in the treatment arms of the time-course of the disease prior to randomization.
Both studies 1–4 enrolled patients with specific molecular profiles (Box 1) rather than the usual unselected populations with metastatic melanoma studied hitherto in phase II or III trials. Nevertheless the activity of Ipilimumab appears to be independent of HLA status.
The illustrated responses achieved with Plexxikon in the selected patients are indeed stunning. 3
However medical oncologists engaged in the treatment of melanoma are familiar with such dramatic responses. Large tumour volumes from widely metastatic melanoma can regress partially and very occasionally completely with treatments combining drugs such as vinca alkaloids, platinum analogues 16–18 and more recently taxanes. 19–21 Such responses are reproducible in 20–30% of unselected patients and have been documented in the literature for over 30 years. Although such responses have not been shown in randomized prospective trials to impact on survival, individual patients can occasionally survive in remission well beyond five years and exceptionally very much longer. 22,23 It is tempting to speculate whether responding patients to conventional chemotherapy harbour the V600E B-RAF mutation and this may be an area of future investigation.
It is estimated that in an unselected population of patients with metastatic melanoma the response rate with Plexxikon is expected to be of the order of 40%, perhaps not significantly higher than the response rates achievable with conventional chemotherapy. This does not detract from the fact that such responses with Plexxikon are possible with a pill taken twice daily rather than with the tribulations associated with intravenous chemotherapy.
It is interesting that Sorafenib, an inhibitor of B-RAF, both of the wild type and of the V600E mutant, exhibited no significant activity against melanoma in clinical trials 3,24–26 for reasons that are not yet clear.
Equally important is the observation that another B-RAF inhibitor GSK2118436 studied in an Australian trial exhibited significant activity against asymptomatic cerebral metastases from melanoma. 27 Extension of this study to a larger number of patients with symptomatic cerebral involvement is awaited with interest. Notably, resolution of cerebral lesions in the Australian trial was seen concomitantly with responding extra-cerebral metastases; such concomitant response to conventional chemotherapy of symptomatic cerebral and extra-cerebral melanoma-metastases has been documented since the early 1980s. 17,28
Intensive research is now focusing on the mechanisms underlying both the sensitivity and the development of resistance to RAF inhibitors of melanomas with the BRAF V600E mutation. 29
Oncologists treating melanoma, starved of innovation for years, will eagerly embrace the new approaches discussed above. Time will tell whether these novel treatments will find meaningful application at the bedside but there is little doubt that for melanoma the era of individualised therapy, based on molecular targets has begun. The ‘penicillin moment for cancer’ has yet to arrive but, for melanoma it may have reached an instant closer.
DECLARATIONS
Competing interests
None declared
Funding
None
Ethical approval
Not applicable
Guarantor
SR
Contributorship
SR is the sole contributor
Acknowledgements
None