Initial treatment for adults with advanced kidney cancer (renal cell carcinoma)


• renal cell carcinoma (RCC)

• avelumab (AVE)

• axitinib (AXI)

• cabozantinib (CAB)

• ipilimumab (IPI)

• lenvatinib (LEN))

• nivolumab (NIV)

• pazopanib (PAZ)

• pembrolizumab (PEM)

• sunitinib (SUN)

Key messages

• When making treatment decisions, it is important to think about whether drugs lengthen life, and whether they decrease or increase harmful side effects.

• The findings in this review apply mostly to advanced renal cell carcinoma (RCC) with a clear cell component.

What is advanced RCC, and how is it treated?

RCC is a type of kidney cancer. It is more common in older people and in men than in women. This is because age (≥60 years) and male sex put people at higher risk of getting it. Other risk factors include body weight, smoking, a history of kidney stones and high blood pressure. More than half of people with RCC discover they have it from routine health check-ups, because many do not have symptoms in the early stages. When symptoms appear, they can impact people's quality of life and day-to-day activities. Before 2005, drugs for treatment of advanced RCC were few and treatments caused many side effects. Now, there are new types of drugs: immunotherapy (use people’s own immune system to find and destroy cancer cells), or targeted therapy (interferes with molecules that are responsible for helping cancer cells to grow, divide, and spread). Combinations of these drugs are used for therapy. With these drugs, people may live longer, with a good quality of life and fewer or milder side effects. These drugs are evaluated in clinical studies with people with RCC.

What did we want to find out?

We wanted to use the most up-to-date information from clinical studies to measure the benefits and harms of different treatments for people with advanced RCC. We also wanted to learn if the drugs worked better for some people than others. 

What did we do?

We searched for studies that explored different drugs that are immunotherapies or targeted therapies. We examined these in adults (≥18 years) with advanced RCC who receive their first therapy. We compared these drugs to the drug SUN, which is a widely used targeted drug and a commonly used comparator drug in studies. We used a standardised process to assess the quality of the findings and our certainty in them. We rated our certainty in the findings based on factors such as study methods, the number of participants in them, and the precision of study results. 

What did we find?

We found 36 studies with 4116 women and 11,061 men, around 60 years of age, with advanced RCC. Most people had ≥2 metastatic sites. We found 22 drugs and 17 combinations of drugs that were measured in the studies. We also performed analyses for different risk groups of advanced RCC. We present and discuss our results for the different risk groups, drugs and combinations in the main text of this review, plus further outcomes. Below we present our main results for our primary outcomes, when all risk groups are combined. We focus on selected drugs (and combinations) (PEM+AXI, AVE+AXI, NIV+CAB, LEN+PEM, NIV+IPI, CAB alone, PAZ alone) that are currently recommended in international guidelines for the treatment of advanced RCC. We report their impact on survival, quality of life and serious side effects.

How long do people live?

People live an average of 28 months when treated with SUN. In comparison, people may live an average of 43 months with LEN+PEM, probably 41 months with NIV+IPI, probably 39 months with PEM+AXI, and probably 31 months with PAZ alone. We are uncertain whether people live an average of 34 months with CAB alone. We do not have information for AVE+AXI and NIV+CAB.

How do people rate their quality of life?

People who receive PAZ alone reported a higher level of quality of life than people who receive SUN, but we are uncertain about the findings. We do not have information for PEM+AXI, AVE+AXI, NIV+CAB, LEN+PEM, NIV+IPI or CAB alone. 

What is people's risk for serious side effects?

People who receive SUN have an average risk of 40% for experiencing serious side effects. In comparison, the average risk is probably: 61% with LEN+PEM, 57% with NIV+IPI, 52% with PEM+AXI, and 40% with PAZ. We are uncertain whether the risk is on average 37% with CAB alone. We do not have information for AVE+AXI and NIV+CAB.

What are the limitations of the evidence?

More studies are needed where these new drugs (and combinations) are not only compared to SUN alone, but also to each other. We lack information on the comparative benefits and harms of these drugs in different people, e.g. when comparing men with women, or different histology types of RCC (e.g. clear cell type, papillary type, sarcomatoid type).

How up to date is this evidence?

We conducted our last search for studies in February 2022 and incorporated the most recent study results into this review. 

Authors' conclusions: 

Findings concerning the main treatments of interest comes from direct evidence of one trial only, thus results should be interpreted with caution. More trials are needed where these interventions and combinations are compared head-to-head, rather than just to SUN. Moreover, assessing the effect of immunotherapies and targeted therapies on different subgroups is essential and studies should focus on assessing and reporting relevant subgroup data. The evidence in this review mostly applies to advanced clear cell RCC.

Read the full abstract...

Since the approval of tyrosine kinase inhibitors, angiogenesis inhibitors and immune checkpoint inhibitors, the treatment landscape for advanced renal cell carcinoma (RCC) has changed fundamentally. Today, combined therapies from different drug categories have a firm place in a complex first-line therapy. Due to the large number of drugs available, it is necessary to identify the most effective therapies, whilst considering their side effects and impact on quality of life (QoL).


To evaluate and compare the benefits and harms of first-line therapies for adults with advanced RCC, and to produce a clinically relevant ranking of therapies. Secondary objectives were to maintain the currency of the evidence by conducting continuous update searches, using a living systematic review approach, and to incorporate data from clinical study reports (CSRs).

Search strategy: 

We searched CENTRAL, MEDLINE, Embase, conference proceedings and relevant trial registries up until 9 February 2022. We searched several data platforms to identify CSRs.

Selection criteria: 

We included randomised controlled trials (RCTs) evaluating at least one targeted therapy or immunotherapy for first-line treatment of adults with advanced RCC. We excluded trials evaluating only interleukin-2 versus interferon-alpha as well as trials with an adjuvant treatment setting. We also excluded trials with adults who received prior systemic anticancer therapy if more than 10% of participants were previously treated, or if data for untreated participants were not separately extractable.

Data collection and analysis: 

All necessary review steps (i.e. screening and study selection, data extraction, risk of bias and certainty assessments) were conducted independently by at least two review authors. Our outcomes were overall survival (OS), QoL, serious adverse events (SAEs), progression‐free survival (PFS), adverse events (AEs), the number of participants who discontinued study treatment due to an AE, and the time to initiation of first subsequent therapy. Where possible, analyses were conducted for the different risk groups (favourable, intermediate, poor) according to the International Metastatic Renal-Cell Carcinoma Database Consortium Score (IMDC) or the Memorial Sloan Kettering Cancer Center (MSKCC) criteria. Our main comparator was sunitinib (SUN). A hazard ratio (HR) or risk ratio (RR) lower than 1.0 is in favour of the experimental arm.

Main results: 

We included 36 RCTs and 15,177 participants (11,061 males and 4116 females). Risk of bias was predominantly judged as being 'high' or 'some concerns' across most trials and outcomes. This was mainly due to a lack of information about the randomisation process, the blinding of outcome assessors, and methods for outcome measurements and analyses. Additionally, study protocols and statistical analysis plans were rarely available.

Here we present the results for our primary outcomes OS, QoL, and SAEs, and for all risk groups combined for contemporary treatments: pembrolizumab + axitinib (PEM+AXI), avelumab + axitinib (AVE+AXI), nivolumab + cabozantinib (NIV+CAB), lenvatinib + pembrolizumab (LEN+PEM), nivolumab + ipilimumab (NIV+IPI), CAB, and pazopanib (PAZ). Results per risk group and results for our secondary outcomes are reported in the summary of findings tables and in the full text of this review. The evidence on other treatments and comparisons can also be found in the full text.

Overall survival (OS)

Across risk groups, PEM+AXI (HR 0.73, 95% confidence interval (CI) 0.50 to 1.07, moderate certainty) and NIV+IPI (HR 0.69, 95% CI 0.69 to 1.00, moderate certainty) probably improve OS, compared to SUN, respectively. LEN+PEM may improve OS (HR 0.66, 95% CI 0.42 to 1.03, low certainty), compared to SUN. There is probably little or no difference in OS between PAZ and SUN (HR 0.91, 95% CI 0.64 to 1.32, moderate certainty), and we are uncertain whether CAB improves OS when compared to SUN (HR 0.84, 95% CI 0.43 to 1.64, very low certainty). The median survival is 28 months when treated with SUN. Survival may improve to 43 months with LEN+PEM, and probably improves to: 41 months with NIV+IPI, 39 months with PEM+AXI, and 31 months with PAZ. We are uncertain whether survival improves to 34 months with CAB. Comparison data were not available for AVE+AXI and NIV+CAB.

Quality of life (QoL)

One RCT measured QoL using FACIT-F (score range 0 to 52; higher scores mean better QoL) and reported that the mean post-score was 9.00 points higher (9.86 lower to 27.86 higher, very low certainty) with PAZ than with SUN. Comparison data were not available for PEM+AXI, AVE+AXI, NIV+CAB, LEN+PEM, NIV+IPI, and CAB.

Serious adverse events (SAEs)

Across risk groups, PEM+AXI probably increases slightly the risk for SAEs (RR 1.29, 95% CI 0.90 to 1.85, moderate certainty) compared to SUN. LEN+PEM (RR 1.52, 95% CI 1.06 to 2.19, moderate certainty) and NIV+IPI (RR 1.40, 95% CI 1.00 to 1.97, moderate certainty) probably increase the risk for SAEs, compared to SUN, respectively. There is probably little or no difference in the risk for SAEs between PAZ and SUN (RR 0.99, 95% CI 0.75 to 1.31, moderate certainty). We are uncertain whether CAB reduces or increases the risk for SAEs (RR 0.92, 95% CI 0.60 to 1.43, very low certainty) when compared to SUN. People have a mean risk of 40% for experiencing SAEs when treated with SUN. The risk increases probably to: 61% with LEN+PEM, 57% with NIV+IPI, and 52% with PEM+AXI. It probably remains at 40% with PAZ. We are uncertain whether the risk reduces to 37% with CAB. Comparison data were not available for AVE+AXI and NIV+CAB.