Dowlati Compares the Advantages and Disadvantages of Atezolizumab and Durvalumab in ES-SCLC

During a Targeted Oncology case-based roundtable event, Afshin Dowlati, MD, discussed the data supporting the use of atezolizumab and durvalumab as therapy for patients with small cell lung cancer.

Targeted OncologyTM: What are the recommended primary therapy options for patients with ES-SCLC?

DOWLATI: Based on the NCCN [National Comprehensive Cancer Network] guidelines, 4 cycles of chemotherapy are generally recommended, but some patients may receive up to 6 cycles of therapy.1 I don’t think I’ve done 6 cycles in the past 10 years.

The preferred regimens are the following: carboplatin [Paraplatin], etoposide [VePesid, Toposar], atezolizumab [Tecentriq] followed by atezolizumab maintenance; or carboplatin, etoposide, plus durvalumab [Imfinzi] followed by durvalumab maintenance. For durvalumab, the indication also includes cisplatin [Platinol], while for atezolizumab, it’s usually only with carboplatin.

The other regimens for patients who cannot get an immune checkpoint inhibitor [ICI] would just be carboplatin or cisplatin plus etoposide, and in rare circumstances, carboplatin, or cisplatin plus irinotecan [Camptosar]. For example, I’ve got a patient who had a massive reaction to etoposide and almost got intubated. I’m treating them with carboplatin plus irinotecan, but such cases should be the exceptions rather than rules. So most patients would be [treated with] carboplatin and etoposide with an ICI, or without in certain circumstances.

Do the data support the addition of the ICI atezolizumab to platinum-etoposide chemotherapy?

Based on data from the IMpower133 trial [NCT02763579], there’s a couple [of points that contrast with those from] the CASPIAN study [NCT03043872]. In IMpower133, there were about 200 patients per arm with a good performance status of 0 to 1, and no prior systemic therapy for ES-SCLC. Patients with treated asymptomatic brain metastases were eligible, but untreated patients were not eligible. It doesn’t mean one can’t use the regimen, but the study did not allow patients with untreated asymptomatic brain metastases.

They went on to receive carboplatin, etoposide, [and] atezolizumab or carboplatin plus etoposide alone. Cisplatin was not part of the study. So they got the carboplatin plus etoposide standard of 4 cycles with or without atezolizumab followed by maintenance atezolizumab or placebo. The other important point about the IMpower133 study is the coprimary end points were overall survival [OS] and an investigator-assessed progression-free survival [PFS], so they had 2 primary end points and wanted to see improvements in both PFS and OS.2

The median PFS in the atezolizumab arm was 5.2 months and 4.3 months in the placebo arm.2,3 One of our colleagues [mentioned that their impression with the treatment] has been a PFS of around 9 months, but the trial [data] suggest a PFS of 5.2 months [with the stratified hazard ratio for disease progression or death at 0.77 (95% CI, 0.62-0.96; P = .02)]. This is a less-than-1-month improvement in PFS. I think that for someone like me who treats about 100 patients with SCLC a year, improvements in PFS don’t mean much.

What we want to see are improvements in OS, [because] I want to see people live longer. That’s the key part about SCLC, not only living longer but having longer-term survival if possible. In this case, the P value is statistically significant, but a less-than-1-month improvement in PFS is nothing to brag about. The OS is much more important. It is 12.3 months for the atezolizumab arm and 10.3 months for the placebo arm. We wanted to see improvements in OS and, indeed, there’s a 2-month improvement in OS and the updated hazard ratio is 0.76 [95% CI, 0.60-0.95; P = .0154].

Now, the other important part about this study is they updated the results just about 1.5 years ago. [Moreover,] there’s about a 13% difference in 18-month survival rates and at 24 months there’s little difference between the 2 arms, but they don’t have data for longer than 24 months of follow-up.3

What were the results of the subgroup analyses?

It seemed [from the analyses] that every single category of patients benefited from the addition of immunotherapy to their treatment. However, upon closer inspection, 2 groups created a little bit of a concern and the first is [patients with] brain metastases. Only 35 patients with brain metastases were enrolled and the 95% confidence intervals were large at [0.46-2.01]. The hazard ratio was 0.96, so it remains questionable at this point based on IMpower133 [findings whether] these patients benefit from the addition of immunotherapy.

The other group that’s a little concerning are patients younger than 65 years old.3 Now, we don’t tend to put too much weight on subgroup analyses—oncologists have been trained to not give too much attention to subgroup analyses— but this is 200 patients, and there’s a little concern for younger patients [HR, 0.94; 95% CI, 0.68-1.28].

So there’s the question in a couple of groups of whether they are benefiting from ICIs. Now, there’s another issue that’s extremely interesting that many may not have paid attention to. There were 200 patients per arm, and when you add pembrolizumab [Keytruda] to carboplatin and pemetrexed [Alimta] in non–small cell lung cancer [NSCLC], the response rates go up. We were hoping that with the addition of atezolizumab to carboplatin and etoposide, we would see a higher response rate but we’re not seeing that. The placebo arm with carboplatin plus etoposide alone had a response rate of 64%.2,4 The atezolizumab arm had a response rate of 60%. So not only is the PFS barely making a 1-month improvement, and there’s some subgroup analysis issues in patients less than 65 years and those with brain metastases, but we’re not even seeing an increase in response rate with the addition of immunotherapy.

For those of us who delve into these details, especially with SCLC being my world, this was a little concerning. I questioned if it is worth using atezolizumab with the IMpower133 data. I used it, but deep down I was questioning how much value it had; but because the survival did improve substantially, it is still a good treatment to give our patients. Now, the good news is that in terms of adverse events [AEs], there’s not much increased toxicity. There’s no increase in neutropenia, anemia, etc. However, we saw the classic immune-related AEs, but other than that no enhanced toxicity, so [it’s] a well-tolerated regimen.2

Do the data support the addition of the ICI durvalumab to platinum-etoposide chemotherapy?

The phase 3 CASPIAN study had 3 arms…[and] I will focus on the platinum plus etoposide with or without durvalumab arms, which could be cisplatin or carboplatin. For the etoposide plus a platinum alone arm, one could go up to 6 cycles, so the placebo arm allows you to give more chemotherapy, whereas with durvalumab, one is stuck at 4 cycles followed by durvalumab maintenance.

On the etoposide and platinum arm, patients also had the option of getting PCI [prophylactic cranial irradiation].5 There are some differences with the IMpower133 study, such as that patients in the control arm could go to 6 cycles of treatment, so if you believe 6 is better than 4, then there’s an advantage. It also allowed for optional PCI, whereas the atezolizumab study did not, and patients with asymptomatic brain metastases were allowed in the study. Finally, the primary end point here was purely OS alone, not together with PFS, and it was a much larger study of close to 300 patients per arm, not 200.

The IMpower133 study had a 24-month follow-up where the curves were getting very close. The CASPIAN study had a 3-year follow-up with about a 12% difference in OS at 3 years. So not only is the median OS very similar to the atezolizumab study, the CASPIAN study improved OS by 2.4 months [HR, 0.71; 95% CI, 0.60-0.86; P = .0003]. The hazard ratio in the IMpower133 study was 0.79. For CASPIAN, it’s a little lower but statistically significant. For the 3-year survival, 13 patients were alive on the etoposide and platinum arm [and] 39 were alive on the etoposide, platinum, and durvalumab arm.6 That’s a 3-fold increase.

What other data support this approach?

There’s a 3-times-higher chance of being alive at 3 years compared [with] chemotherapy alone. This is one of the things that I discuss with my patients, in addition to improvements in quality of life, improvements in OS, and improvements in response rate. We would have never imagined a close to 18% 3-year survival with ES-SCLC, so these are some landmark data.

The subgroup analysis is exactly like the [findings from the] IMpower133 study. The first impression is that every group is benefiting just like in the IMpower study. Instead of 35 patients like for IMpower133, there’s 55 patients with asymptomatic untreated brain metastases. The hazard ratio is not 0.94, like for IMpower133, but a little better at 0.76 [95% CI, 0.43-1.33]. Every age group benefited from the addition of durvalumab, unlike in IMpower133.6 The median PFS is also similar in both arms at 5.1 months for the durvalumab arm and 5.4 months for the chemotherapy alone arm.7 We know from melanoma studies that PFS may not improve but OS improves substantially, so I bring this up because PFS is not an appropriate primary end point for SCLC frontline therapy.

Pembrolizumab is not approved as first-line treatment for SCLC, and the reason because they had a PFS end point, which it did not meet, and therefore the study got rejected at the FDA level. One of the other problems we had with the IMpower133 study is not seeing an increase in response rate with the addition of atezolizumab to chemotherapy.

This is troublesome to us because in NSCLC, we saw increases in response rate. In the CASPIAN study, we saw increases in response rate of about 10% with the addition of immunotherapy. Single-agent immunotherapy has about a 10% response rate in SCLC, and that 10% is nicely added when combined with chemotherapy, but the duration of response was 5.1 months in both arms.6

What was the safety profile of these treatments?

The AEs were similar, and my personal impression is I’m not seeing as much pneumonitis in SCLC compared with NSCLC [Table6]. That’s just my personal experience. When I started using atezolizumab and then durvalumab came out, I put these side by side over time. The durvalumab study has a couple of differences. It can be given with etoposide plus cisplatin or carboplatin, but for atezolizumab it is with etoposide plus carboplatin. The median follow-up in the CASPIAN study is about 1 year longer than the median follow-up in the IMpower133 study, even though the IMpower study [data were] published first.

The arms in the IMpower studies had 4 cycles of chemotherapy. The control arm in the CASPIAN study got 6 cycles of chemotherapy. The median OS was 2 months better on the IMpower study and 2.4 months on the CASPIAN study. The hazard ratio for the IMpower study was 0.76, and it was a bit better on the CASPIAN study at 0.71. The PFS has little importance, and the overall response rates showed a 10% increase in the CASPIAN study. With the CASPIAN study, grade 3 toxicities and immune-related AEs are similar.2-7

We used to use atezolizumab every 3 weeks and durvalumab was every 4 weeks, so there’s some dosing benefit. But all those have been resolved because we now can use atezolizumab every 4 weeks, so I don’t think there’s any administration issues that make any difference between these regimens.


1. NCCN. Clinical Practice Guidelines in Oncology. Small cell lung cancer, version 1.2023. Accessed September 13, 2022.

2. Horn L, Mansfield AS, Szczęsna A, et al; IMpower133 Study Group. First-Line atezolizumab plus chemotherapy in extensive-stage small-cell lung cancer. N Engl J Med. 2018;379(23):2220-2229. doi:10.1056/NEJMoa1809064

3. Liu SV, Reck M, Mansfield AS, et al. Updated overall survival and PD-L1 subgroup analysis of patients with extensive-stage small-cell lung cancer treated with atezolizumab, carboplatin, and etoposide (IMpower133). J Clin Oncol. 2021;39(6):619-630. doi:10.1200/JCO.20.01055

4. Reck M, Liu SV, Mansfield AS, et al. IMpower133: updated overall survival (OS) analysis of first-line (1L) atezolizumab (atezo) + carboplatin + etoposide in extensive-stage SCLC (ES-SCLC). Ann Oncol. 2019;30(suppl 5):v710-v717. doi:10.1093/annonc/mdz264

5. Paz-Ares L, Dvorkin M, Chen Y, et al; CASPIAN Investigators. Durvalumab plus platinum-etoposide versus platinum-etoposide in first-line treatment of extensive-stage small-cell lung cancer (CASPIAN): a randomised, controlled, open-label, phase 3 trial. Lancet. 2019;394(10212):1929-1939. doi:10.1016/S0140-6736(19)32222-6

6. Paz-Ares L, Chen Y, Reinmuth N, et al. Durvalumab, with or without tremelimumab, plus platinum-etoposide in first-line treatment of extensive-stage small-cell lung cancer: 3-year overall survival update from CASPIAN. ESMO Open. 2022;7(2):100408. doi:10.1016/j.esmoop.2022.100408

7. Goldman JW, Dvorkin M, Chen Y, et al; CASPIAN Investigators. Durvalumab, with or without tremelimumab, plus platinum-etoposide versus platinum-etoposide alone in first-line treatment of extensive-stage small-cell lung cancer (CASPIAN): updated results from a randomised, controlled, open-label, phase 3 trial. Lancet Oncol. 2021;22(1):51-65. doi:10.1016/S1470-2045(20)30539-8