Addition of gemcitabine at the time of sunitinib resistance in metastatic renal cell cancer


Shuchi S. Pandya, MD, Beth Israel Deaconess Medical Center, 330 Brookline Ave, East Campus, Boston, MA 02215, USA. e-mail:


Study Type – Therapy (case series)

Level of Evidence 4

What’s known on the subject? and What does the study add?

Second line therapy in metastatic RCC is often mTOR directed therapies. There are no approved drugs to overcome VEGFR tyrosine-kinase inhibitor (TKI) resistance. The combination of gemcitabine and sunitinib has been a safe and active combination in untreated mRCC based on Phase I data. However, the benefit of adding gemcitabine to sunitinib in sunitinib refractory cases was unknown until our case series.

This case series provides clinical rationale for combining sunitinib and gemcitabine in sunitinib refractory cases. This novel combination could be formally tested in a prospective randomized trial to determine its activity and impact on progression free survival. Additionally, further testing could provide insight into the mechanisms involved in overcoming sunitinib resistance.


• To investigate the role of adding gemcitabine at the time of sunitinib failure in patients with metastatic renal cell cancer (mRCC).


• A retrospective chart review of 13 patients with mRCC who experienced disease progression on sunitinib and received gemcitabine 750 mg/m2 i.v. on days 1 and 8 and sunitinib 37.5 mg/day on days 2–15 every 21 days from September 2006–2009 was carried out.

• Responses were assessed using the Response Evaluation Criteria in Solid Tumors.

• The median age was 59, male to female ratio 10:3, Eastern Cooperative Oncology Group performance status 1–3; the number of prior regimens ranged from one to five, and duration on sunitinib alone ranged from 3 months to >1 year.


• Twelve out of 13 patients received more than two cycles with one documented partial response, five stable disease and seven progressive disease.

• The median time to progression was 21 weeks.

• We documented grade 4 pulmonary emboli (n= 1), grade 3 neutropenia (n= 2), anaemia (n= 2) and thrombocytopenia (n= 1).


• The combination of sunitinib and gemcitabine in patients with mRCC may delay disease progression in some patients exhibiting resistance to sunitinib. This regimen merits further prospective investigation in this patient population.


metastatic RCC


vascular endothelial growth factor (receptor)


mammalian target of rapamycin


tyrosine kinase inhibitor


Response Evaluation Criteria in Solid Tumors


Eastern Cooperative Oncology Group


receptor activator of nuclear factor κB (ligand).


In 2009, the estimated new cases of RCC in the USA rose to 57 760 and were associated with 12 980 expected deaths [1]. Available therapeutic options for metastatic RCC (mRCC) have expanded over the past decade with the development of agents targeting vascular endothelial growth factor receptors (VEGFRs) and the mammalian target of rapamycin (mTOR) [2,3]. Sunitinib is a multi-targeted tyrosine kinase inhibitor (TKI) with activity against VEGFR 1, 2, 3, platelet-derived growth factor receptor, Flt3 and c-KIT which was approved by the US Food and Drug Administration (FDA) in 2006 for advanced RCC and has become a standard first-line therapy for patients with RCC [4–6]. Although sunitinib and other similarly targeted agents represent a breakthrough in RCC therapy, durable or complete responses are rarely reported and the vast majority of patients will experience disease progression.

Recently, everolimus, an oral inhibitor of mTOR, showed significant prolongation of progression-free survival vs placebo (4.0 vs 1.9 months; P < 0.001) in a randomized phase 3 trial in patients with advanced RCC who had failed VEGF-targeted TKI [3]. Based on these findings, everolimus was FDA approved and may be considered a standard therapeutic option following progression on VEGF-targeted TKI therapy. Despite these results, the development of alternative therapeutic strategies for TKI-resistant patients remains an area of active clinical and translational investigation. One consideration has been the maintenance of blockade of VEGF signalling following progression on VEGF-targeted TKI. Treatment with agents such as sorafenib and sunitinib have been shown to result in elevated levels of circulating VEGF [4,5]. Upon sudden cessation of VEGFR blockade with sunitinib, acceleration of tumour growth is sometimes observed clinically in the setting of elevated circulating VEGF levels.

Although cytotoxic chemotherapy has modest therapeutic benefit in mRCC, gemcitabine-based combination regimens for rapidly progressive and sarcomatoid mRCC have showed some activity [7–9]. Furthermore, it has been proposed that anti-angiogenic agents, such as sunitinib, may enhance the delivery and efficacy of cytotoxic chemotherapies. Based on observed safety and efficacy in a phase 1 trial of the combination gemcitabine and sunitinib in untreated patients with mRCC, we explored whether the addition of gemcitabine at the time of sunitinib resistance would delay further disease progression in patients with mRCC who were not otherwise eligible for clinical trials [10].


We retrospectively evaluated 13 heavily pre-treated patients with symptomatic or radiographic disease progression while on sunitinib who received gemcitabine at a dose of 750 mg/m2 i.v. over 90 min on days 1 and 8 and oral sunitinib 37.5 mg/day on days 2–15 over a 21-day cycle. Doses of gemcitabine and sunitinib were either held or reduced in the setting of grade 3 myelosuppression based on the National Cancer Institute Common Terminology Criteria for Adverse Events. Evaluated patients completed at least two cycles of the combination. The investigators assessed responses with CT scans using the Response Evaluation Criteria in Solid Tumors (RECIST). Two CT scans separated by at least 6 weeks were required to define stable disease and partial response based on RECIST. The Committee on Clinical Investigations at Beth Israel Deaconess Medical Center granted Institutional Review Board approval for this retrospective chart review.

All 13 patients with clinical or radiographic disease progression on sunitinib were evaluated on the combination of sunitinib and gemcitabine. Ten patients were male, median age was 59 years, median of three prior regimens, and seven patients had an Eastern Cooperative Oncology Group (ECOG) performance status of 2 or 3 prior to initiating this combination (Table 1). The median duration on sunitinib prior to the addition of gemcitabine was 8 months. All patients had prior nephrectomies and clear cell histology without sarcomatoid differentiation, and eight patients had high grade tumours. Eleven patients had two or more sites of metastatic disease with pulmonary disease predominating. Nine patients received prior VEGF-targeted therapies in addition to sunitinib. Seven patients received prior immunotherapy (Table 1).

Table 1.  Baseline characteristics at start of sunitinib and gemcitabine (laboratory normal ranges in parentheses)
PatientECOG performance statusHistologyPrior IL-2 therapy and responsesWBC (K/µL) (4–11)Haemoglobin (ng/dL) 09(14–18)Platelet (K/µL) (150–440)LDH (IU/L) (94–250)Calcium (mg/dL) (8.4–10.2)
  1. IL, interleukin; PEG-IFN, pegylated interferon; WBC, white blood cells; LDH, lactate dehydrogenase.

 13Clear cellNo9.99.511842968.6
 22Clear cell2 cycles low dose IL-2: stable disease for 8 weeks8.510.14667979.2
 32Clear cell2 cycles high dose IL-2: partial response for 44 weeks3.311.61452398.7
 41Clear cellNo4.69.71513348.8
 51Clear cellNo10.314.13331869.4
 61Clear cell2 cycles low dose IL-2: stable disease for 20 weeks5.410.135321610.4
 71Clear cell6 cycles PEG-IFN/IL-2: partial response for 16 weeks7.78.3392
1 cycle high dose IL-2: progressive disease at 2 weeks
 82Clear cell2 cycles high dose IL-2: progressive disease at 7 weeks7.414.5443
 91Clear cellNo4.510.72862879.6
103Clear cellNo2.512.921114610.2
113Clear cell1 cycle high dose IL-2: progressive disease at 8 weeks9.410.53651609.1
120Clear cell1 week high dose, 1 week low dose IL-2: progressive disease at 4 weeks4.913.033710.3
132Clear cellNo2.511.61122138.9


Twelve of the 13 patients evaluated completed at least two cycles of sunitinib and gemcitabine. One patient received fewer than two cycles due to recurrent lower extremity oedema and a diffuse rash which was attributable to gemcitabine. Of the 13 patients in our series, one had a confirmed partial response lasting 21 weeks, five had stable disease for a median of 33 weeks, and seven progressed at a median of 12 weeks (Fig. 1). Five patients experienced longer progression-free survival on the combination than on sunitinib alone (Fig. 2). The overall median time to progression was 21 weeks. The single partial responder experienced a 38% reduction in disease burden and remained on this combination for 58 weeks until disease progression (Fig. 3). All five patients with stable disease had a decrease in disease burden that did not meet the criteria for partial response (Fig. 4). Six of seven patients with progressive disease developed a new metastatic lesion despite relatively stable measurements of residual disease (Fig. 1). Two of these patients developed new CNS metastases at 6 and 21 weeks respectively. Another two patients with established CNS disease developed new CNS lesions at 9 and 12 weeks. Two patients with stable disease developed new bone metastases at 22 and 50 weeks respectively.

Figure 1.

Graphic display of responses. Percentage change in tumour measurements over time (weeks) on sunitinib alone (week <0) and after the addition of gemcitabine (at week 0). *New disease: CNS (n= 4), pulmonary (n= 1), liver (n= 1).

Figure 2.

Horizontal bar graph of progression free survival on sunitinib and gemcitabine compared with prior sunitinib treatment.

Figure 3.

CT scan of partial response. Disease progression on sunitinib at 32 weeks (A) and partial response documented on sunitinib and gemcitabine at 26 weeks (B).

Figure 4.

Waterfall plot of best response on sunitinib and gemcitabine (*patient 3 received fewer than two cycles secondary to gemcitabine related toxicity).

Treatment-related grade 2 toxicities included nausea, vomiting and fatigue. One patient developed grade 2 rash and oedema after one cycle which resolved on stopping gemcitabine and sunitinib. Six patients developed grade 2 or 3 myelosuppression without febrile neutropenia. One patient developed asymptomatic pulmonary emboli. There were no other grade 4 events.


Choosing a second- and third-line treatment for mRCC in the era of VEGF-targeted therapies remains a challenge for clinicians. Based on our described experience, we conclude that the addition of gemcitabine to sunitinib in patients with mRCC may delay disease progression in some heavily pre-treated patients exhibiting resistance to sunitinib. These results are more impressive when one considers that the patients described in this report were generally heavily pre-treated with marginal performance status and not eligible for enrolment on clinical trials. Although there are several caveats in the interpretation of a retrospective observational series, it is striking that the median time to progression of 21 weeks on the combination of sunitinib and gemcitabine is comparable with that obtained with second-line mTOR directed therapy [3]. Furthermore, our series suggests a possible beneficial role of continuing inhibition of VEGF signalling despite the development of disease progression on agents targeting this pathway.

Overall, the pattern of systemic disease progression on sunitinib alone was similar to the pattern of disease progression on sunitinib and gemcitabine. All patients with progressive disease on sunitinib and gemcitabine developed a new metastatic lesion despite stable measurements of residual disease. This observation is consistent with previous reports suggesting that VEGF-targeted therapies are more effective at controlling existing sites of disease rather than preventing new metastases [11]. In this series, the pattern of early disease progression on the combination was predominantly CNS relapses with two patients developing new CNS disease and another two developing progression of established CNS disease. This observation could in part be related to absent or limited CNS penetration of gemcitabine in addition to the natural history of this disease. Two patients with stable disease developed new bone metastases which is a common site of metastatic disease spread in 30% of patients with mRCC [12]. Metastases to the brain and bone may be promoted by other signalling pathways which are not inhibited by sunitinib. Activated STAT3 promotes the overexpression of various molecules including VEGF and matrix metalloproteinase 2 and has been implicated in the development of CNS metastasis in metastatic melanoma [13]. Receptor activator of nuclear factor κB ligand (RANKL) and receptor activator of nuclear factor κB (RANK) are frequently upregulated in bone metastases of patients with prostate cancer. Recently, Mikami et al. [12] showed that elevated RANK and RANKL expression in mRCC correlated with prognosis and development of bone metastases.

There are established precedents for combining cytotoxic chemotherapy and VEGF inhibition in the treatment of chemo-sensitive tumours such as lung, breast and colon cancer. However, in the case of mRCC in which primary VEGF-targeted monotherapy is effective, further prospective investigation is necessary to determine the benefit or superiority of combining chemotherapy with VEGF inhibition. Currently, we are participating in a multicentre phase 2 study designed to assess the activity of sunitinib and gemcitabine in sunitinib-naïve patients with poor risk mRCC. However, mitigating sunitinib resistance with the addition of cytotoxic chemotherapy has not been formally evaluated. A potential study to address this question and validate our findings would be a randomized phase 2 study comparing sunitinib and gemcitabine against mTOR directed therapy in sunitinib refractory patients.

Possible mechanisms for the observed additive effects of gemcitabine on sunitinib therapy are speculative at this time [14]. Of note, none of the tumours’ histologies were the sarcomatoid variant typically responsive to gemcitabine, which favours an alternative explanation of gemcitabine’s observed activity in this series. Many have argued that VEGF-targeted therapies result in pseudonormalization of tumour vasculature, resulting in improved delivery of cytotoxic chemotherapy and therefore enhanced efficacy. A more enticing possibility is that gemcitabine may target specific mechanisms of resistance to sunitinib.

Several studies in animal models investigating tumour-induced immune suppression and angiogenic escape have proposed that immature myeloid cells, characterized by co-expression of CD11b and Gr-1, may play a critical role in the development of resistance to VEGF-targeted therapies [15,16]. Shojaei and Ferrara [15] recently showed that tumours resistant to VEGF-targeted therapy were associated with an increase in CD11b/Gr-1+ infiltrating cells compared with sensitive tumours. These cells augmented vascular proliferation in sensitive tumours while mediating resistance in refractory tumours, possibly by upregulating Bv8 (prokineticin-2), a protein which may promote alternative angiogenesis pathways [15,16]. Interestingly, it was also recently shown that gemcitabine selectively reduced the number of CD11b/Gr-1+ cells in the spleens of tumour-bearing mice from five different cell lines without a deleterious effect on the numbers of CD4+, CD8+ T cells and natural killer cells [17]. Although the human equivalents to these cells have yet to be fully characterized, the possibility that gemcitabine may enhance the efficacy of sunitinib by depleting such cells is worthy of further investigation.

In conclusion, our experience of adding gemcitabine to sunitinib as a salvage regimen in 13 heavily pre-treated and marginal performance status patients exhibiting sunitinib resistance suggests a promising therapeutic strategy for overcoming resistance to VEGF-targeted therapy. However, as this experience has only been observational, further prospective randomized clinical trials are necessary to validate our findings.


No direct funding was obtained for this research.


None declared.