Quite a number of patients do receive concurrent chemoradiation in patients with head and neck squamous cell carcinoma and the problem is that in patients who receive this concurrent chemoradiation nearly 60-70% do develop grade 3 or higher oral mucositis. This oral mucositis significantly impairs not only the patient quality but also the delivery of subsequent therapy as there is significant treatment delay, both in the radiation and the chemotherapy perspective once these patients do develop oral mucositis. Also there is  a lot of malnutrition, the patients become frail. So there is a lot of challenge once these patients develop oral mucositis and the pain associated with oral mucositis.

What was the unmet need and what was the challenge? The challenge was that there is a lack of access to stronger opioids, especially in low- and middle-income countries, especially access to morphine, buprenorphine, fentanyl. If I have to put in some numbers, around 4-5% of these patients actually do receive strong opioids in low- and middle-income countries, so there is a huge unmet need. Secondly, once there is oral mucositis induced pain there is a nociceptive and a neuropathic component to pain and it is in this neuropathic component where actually the efficacy of morphine and other stronger opioids may not be fully there.

So here is where the role comes for gabapentin. Gabapentin has been studied previously in a small subset of patients, there is no randomised study as such. But gabapentin, as such, is widely available, it is not toxic, it does not bind to plasma protein, it is excreted as a whole through the kidneys. So it is a safe drug, widely available, not expensive and prior small studies have shown that gabapentin actually does reduce this oral mucositis pain.

So we thought to take this regime forward, we thought to take this drug, gabapentin, forward and actually test it in a phase III randomised study – can it actually reduce the pain associated with oral mucositis.

What was the study design?

This was, again, a phase III randomised study. It was an open-label superiority designed study where those patients who were receiving chemoradiation, whether it was definitive or adjuvant didn’t matter, but those who were receiving chemoradiation and once they received chemoradiation if they developed oral mucositis-associated pain. So we assessed pain on the visual analogue scale and if the pain was one or more these patients were eligible in the study. We randomised 154 patients to two arms: in arm A patients received a local anaesthetic that is 20% weight by volume benzocaine, along with that they received 15mg of tramadol given twice a day orally. In arm B actually patients received along with the local anaesthetic and tramadol patients received gabapentin in escalating doses starting from 300mg to a maximum of 1800mg.

The primary endpoint of the study was actually to see the area under the curve of the pain scores in the two arms. So the pain scores were assessed at baseline, at eight time points over the next six hours on day 1 and on day 7. The primary endpoint was actually to see what are the area under the curve of the pain scores in both the arms. There were a few secondary endpoints and they were to assess the toxicity between the arms, the quality of life in case there is any delay in the treatment, and compliance. So these were the key secondary endpoint.

One key secondary endpoint was to see if there is any rate of change in analgesia after one week of starting the treatment in both these arms. So that is something called requirement of rescue analgesia and that was also a key secondary endpoint in our study.

What were the key findings?

The study was negative for its primary endpoint. So, as you know, the primary endpoint was to assess the area under the curve in both the arms for the pain scores which were assessed at multiple timepoints – at baseline, on day 1 and day 7. So actually the trend line plots do show that there was a significant decrease in the pain score in both the arms, however, the amount or the quantity of pain score reduction was similar in both the arms. Therefore our primary study endpoint did turn out to be negative.

However, there was a key secondary endpoint of requirement of rescue analgesia and that was statistically significantly better in the gabapentin arm. So around 38-40% of patients in the tramadol arm did require escalation to stronger analgesia, what we call analgesic rescue, whereas only 12% of patients in the gabapentin arm actually required an analgesic rescue, that is escalation to a stronger opioid, most commonly we used morphine in those patients in which the pain score was not controlled. This was statistically significant when we assessed both on a univariable and on a multivariable analysis, showing that there was an 80% reduction in the odds of requiring rescue analgesia when patients actually received gabapentin. That was a key secondary endpoint which turned out to be positive.

The other endpoints, that is treatment delay, quality of life, all of these were statistically similar in both the arms. However, there was a trend towards reduced symptom burden, there was a trend towards better quality of life in those patients receiving gabapentin, however that did not turn out to be statistically significant.

So the key study endpoint is that the reduction in pain score in both the arms were similar, so the AOC, the median and the main AOC in the VAS scores were similar in both the arms. However, statistically significant reduction in the need for rescue analgesia was there in the gabapentin arm.

What could be the clinical significance of thee results?

Both globally and in low- and middle-income countries where there is a shortage in access to these stronger opioids, especially morphine, fentanyl, buprenorphine. So a number of countries do have difficulties in access to these stronger opioids. Especially in these scenarios gabapentin plus tramadol plus a local anaesthetic significantly reduces the pain scores. Not only it reduces the pain scores in those who develop oral mucositis, there’s a significant reduction in the need for rescue analgesia. This actually prevents or saves a number of patients who eventually go on to require stronger analgesics. As we know that stronger analgesics, especially morphine, do have a number of side effects. So, along with the challenge in access to these drugs, these drugs do have a number of side effects as well, whether it’s constipation, whether it’s dependence, whether it’s drowsiness, somnolence, nausea, vomiting. So there are a number of side effects associated with these stronger opioids. By using gabapentin and tramadol there is a significantly lesser number of patients who eventually do end up requiring these stronger analgesics. So there’s a global relevance to this, the combination regimen of local anaesthetic plus tramadol plus gabapentin.

For our patients there was one catch in the study. The catch was that, and this was discussed in ASCO as well, that probably the comparator arm was not a globally relevant comparator arm. So most commonly the regimen that we use globally is platinum-based chemotherapy with full-dose immunotherapy or a platinum-based chemotherapy with cetuximab. The only problem is that this was not a real-world pragmatic comparator because most of the patients cannot access full-dose immunotherapy and, secondly, it adds on to a significant financial burden in actually conducting these studies. Secondly, most of the pharma companies will not sponsor such a study of full-dose versus low-dose immunotherapy, that is why using full-dose immunotherapy or cetuximab as a comparator was a challenge. Thirdly, we  got a comment is that the comparator arm could have been paclitaxel/carboplatin with low-dose immunotherapy but there was a challenge to that as well because it would end up being that both the arms of the study would have been an experimental arm and the study interpretation would not have been valid.

So one of the major questions that we got is why did we select paclitaxel/carboplatin as a comparator and the reason is that paclitaxel/carboplatin is a real-world pragmatic comparator. We hoped that the ideal comparator of paclitaxel/carboplatin and full-dose immunotherapy is something that we could have used but it was really challenging from the financial aspect and we would not have received any sponsorship to do chemotherapy with a full-dose immunotherapy comparator.  Secondly, from a scientific point of view, we couldn’t keep chemotherapy with low-dose immunotherapy as a comparator because chemo with low-dose immunotherapy is not a standard of care globally. So that is why we needed to use a pragmatic, real-world comparator of paclitaxel/carboplatin.