ESMO Webinar with OA Dr. Maximilian Hochmair

• The PACIFIC trial has achieved excellent results using the PD-L1 inhibitor, durvalumab, after simultaneous chemotherapy and radiotherapy – when stage III patients were inoperable.
• The FLAURA trial investigated the use of third generation TKIs in EGFR mutated patients in comparison with first generation TKIs as first line therapy for EGFR mutated patients. The results require further discussion.
• As part of a follow-up, the IFCT-0302 trial looked at how often patients actually require a CT scan after an operation.
• The BRF113928 trial examined dabrafenib and trametinib as first line treatment for BRAF-mutated patients. The results suggest that these substances are now the gold standard for BRAF-mutated patients.
• The CheckMate 153 trial looked at whether it is possible to discontinue immunotherapy after one year.
• And finally, there were also the results of the tumor mutational burden from both the POPLAR and OAK trials. The tumor mutational burden in the blood was determined in these trials, in order to also have a biomarker in the blood, along with the biomarker PD-L1, which is determined in the tissue.

If I may, I shall begin with the PACIFIC trial. In the PACIFIC trial, stage III patients were examined after receiving simultaneous chemotherapy and radiotherapy. We know that approximately one third of patients are stage III. If the patients are inoperable, they often receive simultaneous chemotherapy and radiotherapy. And as we know from our own department, this is sometimes not easy to carry out, but we do at least attempt it. Despite this simultaneous chemotherapy and radiotherapy, the prognosis for these patients is highly unfavourable. The mean PFS in such cases is around eight to ten months, while the five-year survival rate is only about 15 percent. In recent years, there have been virtually no developments for these patient groups.

Durvalumab is a PD-L1 inhibitor which is administered intravenously once every two weeks. The infusion lasts for approximately an hour. You can see from the trial design that in this case patients, who had received at least two cycles of chemotherapy and radiotherapy, were randomised one to 42 days after the end of the simultaneous chemotherapy and radiotherapy. After this, they received durvalumab once every two weeks, or placebo. The randomization was carried out on a 2:1 ratio.

The PFS data currently presented following a BICR show a very positive result. The PFS tripled to 16.8 months versus 5.6 months. This very positive result is clearly statistically significant and suggests that the substance ought to be brought into use as soon as possible.

There were no relevant differences in the sub-group analysis. The only notable trend to appear was that EGFR-mutated patients did not benefit to the same extent from the immunotherapy. The treatment did not appear to be optimal for these patients, although for the other patients it was certainly a very good option. As you can see here, the PD-L1 expression did not play a role. Also, the antitumor activity, that is to say the objective response rate, could be increased from 16.0 to 28.4 percent.

The occurrence of new lesions under durvalumab was clearly reduced in comparison with placebo, and likewise it was possible to delay the time at which new distant metastases and death occurred (23.2 versus 14.6 months). This was a clear advantage of durvalumab, as currently not even half of the patients received the active substance.

In particular, as regards the combination of radiotherapy with immunotherapy, where the immunotherapy led to pneumonitis in 5 percent of patients, it was very interesting to see whether we had an improvement in side effects. However, no changes were found. There were no relevant differences in particular in the area of pneumonitis or radiotherapy-induced pneumonitis.

Across all degrees of severity, side effects occurred in 33.9 % of patients in the durvalumab arm and in 24.8 % in the placebo arm. But for the more common side effects, in particular, there was virtually no difference, with 3.4 % in the durvalumab versus 2.6 % in the placebo arm. Durvalumab is thus a well-tolerated substance. According to the authors, these highly promising data should also be made available to patients as soon as possible.

The FLAURA trial

The second trial which I would like to present to you, concerns the use of osimertinib as a first line therapy for EGFR-mutated patients. It is currently the case that we initially offer a first or second generation TKI to patients with an activated EGFR mutation. If a T790M mutation develops – which happens in approximately 60 to 70 percent of cases – we change to osimertinib: the third generation TKI that also blocks the T790M mutation effectively. The first line data published by Ramalingam et al. in the Journal of Clinical Oncology showed that the substance also led to a very good PFS (22.1 months) in this setting. For this reason, its use as a first line therapy must, of course, be investigated.

In the trial, patients with the EGFR mutation deletion 19 and L858R were randomised in a 1:1 ratio and then received either osimertinib or the standard of care. Regrettably, afatinib was not used, although it is the standard of care in Austria. In this trial, gefitinib and erlotinib were used as the standard therapy. It is also unfortunate that it was not tested whether the sequence of gefitinib and erlotinib, followed by osimertinib, would be better for the T790M mutation. Thus, unfortunately, there was also no fixed arm and we know from the data that hardly any of the patients with the T790M mutation in this standard of care arm received osimertinib.

As regards patient characteristics, there was a typical distribution, even though both Asians and Non-Asians were included in the trial. The PFS showed a clear advantage for osimertinib (18.9 versus 10.2) in the standard of care arm. This is a clearly positive trial, which is clearly statistically significant.

Sub-group analyses also showed that osimertinib was advantageous for all patients in comparison with the standard of care. No group experienced less benefit. Patients with brain metastases showed a similarly good PFS, and we know that osimertinib is highly effective in the brain. In the overall survival rate there was a trend for osimertinib, although the data are still immature and can thus still not yet be used for analysis. In the comparison of adverse events, there was also a clear advantage for osimertinib, as it is more easily tolerated than gefinitib and erlotinib.

In summary, Ramalingam has established that the substance may represent a new standard treatment for EGFR-mutated patients.

In the subsequent discussion, Toni Mok engaged with the points that I raised: neither afatinib nor a second generation TKI were used in the standard of care arm. The crossover from the standard of care to osimertinib was not examined. In addition, there are still no data on re-biopsy or data on treatment after osimertinib. It is therefore still not possible to answer whether the standard of care followed by osimertinib is better for T790M. And, of course, we do not know whether patients without T790M show poorer response. In short, these data do not, in my view, currently alter clinical practice. My sequence still remains afatinib, as we have been able to develop a PFS of up to 14 months, followed by osimertinib, as we know from our own data that up to 70% of patients can still then develop a PFS of 11months.

The IFCT-0302 trial

Now for the third trial that I would like to present to you. The IFCT-0302 trial looked at how frequently we have to treat patients with a follow-up procedure after an operation. To date, it has been our usual practice in line with ESMO guidelines that we carry out a CT scan half-yearly for the first three years and then change to an annual CT scan. However, this has not been tested by means of randomized trials. This trial has a maximum arm and a minimum arm. In the maximum arm, a CT scan was carried out half-yearly in the first three years and thereafter annually. There was also a bronchoscopy for squamous cell carcinomas and large cell cancers. In the minimum arm, an x-ray was carried out half-yearly and then annually after three years. Only when there were symptoms or abnormalities in the x-ray was a CT scan carried out. These were the usual patients, who had been operated on, there were no distinctive features, and no imbalance in the patient groups. The patient cohorts were large with 888 patients in the minimum arm and 887 in the maximum arm. There was no statistically significant difference in the overall survival rate, independently of whether an x-ray was carried out annually or half-yearly and a CT scan only when there were symptoms or abnormalities in the x-ray, or if a CT scan was carried out first automatically. This result is really surprising and we should reconsider our current practice.

As regards disease-free survival, there was a positive trend for the maximum staging, but it was not statistically significant. In a subsequent explorative analysis it was clear, however, that we could detect secondary primary tumors earlier with more frequent CT scans, which could be an advantage for CT scans. The authors’ summing-up, and also my own expectation for the ESMO Guidelines, are that we should move away from six-monthly CT scans and change over to an annual CT scan. In my own day-to-day practice, we carry out half-yearly check-ups and intermittently include a CT scan and x-ray.

BRF113928 trial

The fourth trial tested dabrafenib and trametinib as a first-line therapy for the BRAF mutation. This V600E mutation affects approximately 1.6 percent of patients with non-small cell lung cancer. According to our data, which were presented at this year’s World Conference on Lung Cancer, about 3 percent of the 400 patients examined were affected.

It is already known that this V600E mutation can be treated very well with dabrafenib and trametinib. There are now different cohorts who have already been investigated. At the recent ESMO Cohort C was examined, in which the patients with the V600E mutation received the combination as a first-line therapy. These patients showed a PFS of 14.6 when they received the combination as a first-line therapy in an independent PFS review. We know that these patients rarely respond well to chemotherapy. This is thus a truly excellent PFS for this group with an otherwise poor prognosis.

The overall survival rate was 24.6 months – the mean survival rate for this patient group is normally about 12 months. The response rate is around 64 percent. There was therefore hardly any patient who did not benefit from the combination, who simultaneously had a very long response period.

The combination is very well tolerated. As for side effects, typically fever and other, minor adverse events can occur. Grade 3 or Grade 4 side effects were rare. In short, it is a very well tolerated, highly effective substance in first-line therapy, which should therefore be the gold standard for these patients. There is a recommendation to conduct reflex testing for BRAF, for non-small cell lung cancer, and to then use the substance in the first line. Since August 2017, the substance has already been granted marketing authorization in Austria as a first-line therapy for BRAF-positive patients. It would thus be a good idea to carry out reflex testing and then also to be able to offer patients this substance.

The CheckMate 153 trial

The fifth trial is the CheckMate 153 trial. This examined whether patients could end their treatment after a year after responding to nivolumab, or if the immunotherapy should be continued. We often ask ourselves in our daily practice how long we should treat patients, who respond well to an immunotherapy, with this therapy as the costs of immunotherapy are relatively high. And the patient also frequently asks about how long the treatment will last.

The trial studied patients who had had one year of treatment with nivolumab, and who were in partial or full remission. These patients were either randomized into the group in which the treatment with nivolumab was being continued, or into the group in which therapy was stopped. Nivolumab could be used again after it had been halted in the event of progression. As for PFS, there was a clear advantage in continuing with nivolumab. You can see here that the mean PFS was still not achieved by the patients in the nivolumab arm (10.3 months after a one year’s stop). A clear trend can therefore be seen within the PFS and in the overall survival rate for a non-discontinuation of the substance. However, the data are still too immature to be able to carry out more precise analyses. It was also possible to administer the substance again a year after ending the therapy. The patients did benefit, in part, from the continuation, although some patients gained no advantage from the continuation. The key point of this trial for my patients is therefore that the substance should not be discontinued and thus, as it was investigated, the patients should continue to be treated until progress, until we have more data available.

The POPLAR trial

Lastly, I come to the determination of the tumor mutational burden in the blood. We have a few patients who do not benefit from immunotherapy. We therefore require a biomarker so that we can evaluate which patients might gain greater benefit, and which ones might receive less. In general, immunotherapy has brought enormous advantages for patients. But, despite this, we know of patients who regrettably do not benefit from it.

In the POPLAR trial, the tumor mutational burden in the blood (bTMB) was determined and then validated with the tumor mutational burden blood test in the OAK trial. What is known as the “cut off” for the positive/negative patients is bTMB ≥ 16, in order to evaluate whether the patients will respond better in this case.

In the current PFS data from the OAK trial, there is a clear indication that the bTMB is a predictive marker for the PFS in order to evaluate whether the patients will gain greater benefit from it. The slides and the presentation show here that the PFS is clearly better, if the bTMB is above 16. It can also be seen from the overall survival rate that a higher bTMB is advantageous for patients. The advantage of the bTMB lies in the simplicity of the method, which namely involves taking a blood sample. It also appears to deliver a more homogeneous picture of the overall situation of the tumor in comparison with the tissue tumor mutational burden. Also, no re-biopsy is required.

PD-L1 and bTMB do not correlate with one another. There are patients who have the two, both a high PD-L1 expression and a high bTMB, and these patients respond fantastically well. But there are also patients, who have only the one or the other. With this method, we are hoping to be able to select those patients in future who will respond very well to this therapy, and also that the test will soon be available in Austria.

I hope that you were happy with my selection of trials and also that the clinical background will be useful for you. Thank you all for watching and goodbye!