Siefker-Radtke Compares Chemotherapy and Immunotherapy in Patients With Urothelial Carcinoma

Arlene O. Siefer-Radtke, MD, recently spoke on the treatment considerations and decisions she makes when treating patients with urothelial carcinoma.

Arlene O. Siefer-Radtke, MD

Arlene O. Siefer-Radtke, MD, recently spoke on the treatment considerations and decisions she makes when treating patients with urothelial carcinoma. Radtke, a professor in the Department of Genitourinary Medical Oncology, Division of Cancer Medicine, at The University of Texas MD Anderson Cancer Center in Houston, Texas, explained her treatment decisions based on 2 case scenarios during aTargeted Oncologylive case-based peer perspectives presentation.

Case 1

December 2015

A 68-year-old man wasdiagnosed with clinical stage T4b (locally advanced) urothelial carcinoma of the bladder. He received gemcitabine/cisplatin chemotherapy for 6 cycles with a minimal response and he declined salvage surgery.

Ten months later, he developed progressive disease in the lung. He presented with liver function within normal limits and moderate renal impairment with a creatinine clearance (CrCl) of 49 mL per minute. He also had rheumatoid arthritis that was managed with adalimumab (Humira) and methotrexate. He had an ECOG performance status of 1.

The patient was treated on a clinical trial with durvalumab (Imfinzi) at 10 mg/kg administered intravenously 2 times a week.

Follow-up imaging at 3, 6, and 12 months showed continuing tumor shrinkage.

In April 2016, he presented to the emergency department with symptoms of shortness of breath and cough.

Targeted Oncology: What are your initial impressions of this patient?

SIEFKER-RADTKE:This was an interesting case of a 68-year-old man with a T4b advanced urothelial carcinoma. What T4b tumors mean is that the tumor is invading into the pelvic side wall. There is a designation of T4a, which typically involves invasion of adjacent but resectable organs. But once it is invasive into the pelvic side wall, it indicates a tumor that is fixed and is not surgically resectable upfront. So, it is no longer a neoadjuvant patient.

This patient, who is classified with our stage IV disease, received frontline chemotherapy with gemcitabine and cisplatin for 6 cycles. Unfortunately, he had a minimal response and declined salvage surgery. I believe that this is very appropriate when someone doesn’t have a good response because the bulking procedures on the bladder, where you go in and resect without an attempt at negative margins, don’t typically help patients. They are also associated with very high rates of recurrence and relapse.

Targeted Oncology: What are the second-line treatment options for this patient?

SIEFKER-RADTKE:For patients who fail cisplatin-based chemotherapy, the current standard for eligible patients is an immune checkpoint inhibitor (ICI). There are 5 ICIs currently approved, 2 targeting PD-1 and several targeting PD-L1, which show evidence of clinical activity. In some of them, with longer follow-up, we are seeing the poten­tial for durable response.

Alternatively, single-agent taxanes have been considered the standard, and some patients would get more combination chemo­therapy depending on their overall physical condition. Some may even get more cisplatin, although, arguably, someone who has failed chemotherapy in a relatively short interval and did not have a great response to start with [would probably be] less likely to benefit from more of the typical systemic chemotherapy. I would always favor treating patients with something different and would consider an ICI.

It is interesting that, in this case, the patient was enrolled on a clinical trial. Most of the clinical trials will allow what are considered minor conditions, but patients who are on treatment for rheumatologic conditions are typically excluded from the immunotherapy trials out of concern that the treatment could exacerbate their rheumatoid condition or, alternatively, the immunosuppression that they’re on could actually inhibit the immune response.

Often, when I get a patient who has a rheumatologic condition and they’re on active treatment, among the first questions I ask the patient is: How long have you been on treatment and are you still having active symptoms? There are some patients who have been chronic immunosuppressive for years without reactivation of their condition. In that setting, the next question is whether they can be weaned off the immunosuppression. Intriguingly, that may have an impact on their tumor as well. Bladder cancer is a disease that is associated with immunosuppressive drugs. I have certainly seen many patients who have been on immunosuppres­sion for transplant or rheumatologic conditions who then develop bladder cancer. So, I believe immunosuppressive treatments do contribute to an increased risk of developing this cancer.

If you can wean them off their immunosuppressive drug successfully, then I believe it is certainly reasonable to consider an ICI on a case-by-case approach. You specifically discuss the risks of going on an immune-modulating agent, because this patient would be at an increased risk of having recurrence or worsening of his rheumatoid arthritis. There may be reasons to consider doing this and starting the patient on an ICI because, with his cancer, he has been given an end-stage diagnosis. With other treatments, we are not going to provide durable responses. But if they can get a response to immunotherapy, current data suggest that about 75% of patients who respond remain in response at 1 year, and close to 50% who respond remain in response after 2 years.

Targeted Oncology: What is the rationale for treating the patient with immunotherapy?

SIEFKER-RADTKE:The benefit of immunotherapy, especially when compared with combination chemotherapy, is that it appears to be more toler­able, which is best seen in the randomized data from pembro­lizumab (Keytruda), where they compared the toxicities of pembrolizumab with the toxicities of single-agent taxanes. They saw an improvement in the toxicity profile. One thing that has happened since the approval of ICIs is that we are now seeing groups of patients who were never candidates for chemotherapy that go on to get an ICI. These are patients who are too frail, have poor heart function, or are elderly.

I think these ICIs are being explored in all patients, but there are data suggesting that patients with lymph-node—only disease do better than patients with more visceral metastases. There have been variable data on PD-L1 expression and whether that can predict for response or benefit. While some studies suggest it could be predictive for patients who are more likely to respond, there are additional data suggesting that even patients with low or no PD-L1 expression may respond as well. At the moment, PD-L1 is not a great marker and it is not required for the treat­ment of patients with urothelial cancer.

There are many other strategies currently under development, including looking at tumor mutation burdens with the sense that tumors with a higher mutation burden may respond better than tumors with a low mutation burden. Microsatellite instability is a potential marker, and we’ve seen approvals for many tumors with microsatellite instability to receive pembrolizumab. At the moment, [however,] we don’t have any proven marker where I can test it and tell a patient that they should or should not receive an ICI.

Targeted Oncology: What experience can you expect this patient to have with durvalumab?

SIEFKER-RADTKE:It is certainly an active drug. I think with durvalumab, they looked at PD-L1—high and a combined strategy, which not only looks at the tumor cell but also looks at the immune cells to predict responses. Despite that, while they see trends toward improve­ment in response in the PD-L1–high group, there are still patients with PD-L1–low tumors who respond to durvalumab.

If you look at the response rate for durvalumab, it certainly appears consistent in terms of what we are seeing across the board with ICIs. On average in patients who are post platinum, the objective response rate (ORR) seems to be in the range of 15% to 20% with minor variability, probably related to patient selection. When you look at the durvalumab data specifically, the confirmed ORR overall was around 17%. When they looked at PD-L1—high tumors, it was 27%. However, in the PD-L1–low tumors, the response rate did seem to be lower at 4%.

I would argue that the value of having a predictive marker is not there yet unless we have a range of choices for patients who are trying to select the best one.

Responses to durvalumab have also been shown to be durable, with some responses lasting more than 12 months. Trials are still ongoing to determine the full length and durability of response. As we’ve seen with ICIs, the response may be arguable on the lower side, but the patients who do respond have the potential to have a more durable response than what we’ve seen with systemic chemotherapy.

Targeted Oncology: What are the toxicities associated with durvalumab?

SIEFKER-RADTKE:[In this case,] we have a patient who is presenting with shortness of breath and cough on an ICI. I think it is important for physi­cians in general to know and understand that we cannot ignore what were previously perceived as minor complaints. You cannot just assume, in this case, that [the symptoms are being caused by] an infection, or that they can wait until their next clinic visit.

What I typically tell patients is that if they do have a symptom that could be associated with 1 of the severe toxicities, [we must] take it very seriously. They can progress very quickly and there have been deaths reported with these ICIs in the setting of pneumonitis.

When a patient presents with shortness of breath, we do a workup. This often includes a CT angiogram because you want to see whether they have progressive lung metastases causing the shortness of breath or the cough. We also want to do the CT angiogram to make sure that they have not had a pulmonary embolism, because bladder tumors are among those tumors associated with a higher rate of blood clots as well. While pulmo­nary embolism can have a ground-glass infiltrate in the lungs, we have had patients who have not had much of a ground-glass infiltrate and it appeared very minor. While we don’t have defini­tive diagnostic criteria yet, based on bronchoscopy, we have seen hemosiderin-laden macrophages, which could suggest some leaky vessels in the lungs causing more heme to be taken up by the macrophage groups. We are also looking at CD4:CD8 ratios as a potential predictive marker, but at the moment, we do not have anything definitive where you can say with certainty, yes or no, it is or isn’t pneumonitis.

[We use] a combination of symptoms and clinical suspicion to [try to determine] why the patient is having shortness of breath or coughing. If you think it is pneumonitis, the patient needs to get started on steroids right away to help decrease the pneumonitis and inflammation, and to turn down the immune response. If patients are going to be on steroids, pneumocystis pneumonia prophylaxis should be considered depending on the length of time they have been on steroids.

Case 2

June 2016

A 66-year-old woman was referred by urology after evaluation for gross hematuria. Her past medical history included mention of chronic obstructive pulmonary disease and mild hypertension (although there was no prior myocardial infarction). She has a 40-year smoking history as well.

A cystoscopy was performed and revealed a nodular 8-cm mass along the posterior bladder wall. Transurethral resection of the bladder showed a high-grade urothelial carcinoma invading the muscularis propria. A CT scan of the abdomen and pelvis was negative for lymphadenopathy or distant metastases. She was diagnosed with muscle-invasive bladder cancer, urothelial origin.

Targeted Oncology:How do you consider disease factors and patient factors when deciding on a therapeutic approach?

SIEFKER-RADTKE:Often, we discuss performance status, renal function, and age. A lot of these patients, because they were smokers, have some heart disease and may have an obstructed kidney. Most groups argue that a patient’s EGFR should be 60 mL/min or greater to give full-dose cisplatin. [However,] there are ways of giving full-dose cisplatin to an EGFR of 50 mL/min, or even split-dose cisplatin to an EGFR of 40 mL/min. These patients do need to be monitored closely and you cannot skimp on their hydration if you are giving neoadjuvant chemotherapy.

Targeted Oncology:What is the best approach for this patient?

SIEFKER-RADTKE:For a patient with muscle-invasive bladder cancer, if their EGFR is adequate and they may require a nephrostomy tube if they have an obstructed kidney, the standard approach would be a cispl­atin neoadjuvant chemotherapy regimen, such as DD MVAC. This suggests a response rate downstaging to T0 around 39% based on data from myself, Elizabeth Plimack, and others. We typically give 4 cycles of neoadjuvant DD MVAC. Others are advocating for 3 cycles, but we don’t know if one is better than the other. Additionally, some groups are advocating for gemcitabine plus cisplatin as well.

This patient received neoadjuvant DD MVAC for 4 cycles, had a cystectomy, and at surgery unfortunately did not have a great response. She has a TIIIbN0 high-grade urothelial tumor. When patients have high-stage disease, TIIIb, the risk of recurrence is greater than 50%, probably closer to 70%, suggesting their long-term cure fraction might be in the range of 30%. There is currently no proven role for any additional adjuvant treatment in this setting, although ICIs are being studied in the adjuvant setting. So, if you can get this patient on an adjuvant clinical trial that would be a reasonable strategy. Otherwise, the standard of care would be to proceed with observation.

The patient received neoadjuvant platinum-based chemotherapy (dose-dense methotrexate, vinblastine, doxorubicin, cisplatin; DD MVAC) for 4 cycles and subsequently underwent radical cystectomy in October 2016.

At the time of cystectomy, pathology demonstrated that she had pTIIIbN0 high-grade urothelial carcinoma. She was subsequently placed on observation.

Targeted Oncology:What should be done after the patient progresses on DD MVAC?

SIEFKER-RADTKE:The current approval for good EGFR would be cisplatin-based chemotherapy. However, this patient relapsed within 1 year of receiving cisplatin. Patients who relapse in that time interval typically have a prognosis that is more closely associated with platinum failures. Clinical trials with ICIs have allowed patients who have failed neoadjuvant or adjuvant therapy within 12 months of their last dose of chemotherapy. This patient would be a candidate for an ICI based on the timing of the response.

February 2017

The patient complained of shortness of breath, and a CT scan showed multiple bilateral pulmonary nodules. Metastatic urothelial carcinoma was confirmed by interventional radiology biopsy.

She now presented with liver function tests and CrCl within normal limits and an ECOG performance status of 1. She continues to smoke 1+ packs per day.

Targeted Oncology: Is the patient a candidate for chemotherapy?

SIEFKER-RADTKE:Since we are not curing most of these patients, you could consider chemotherapy. There has been some concern that some patients with rapidly progressive symptoms may not do as well on ICIs as compared with chemotherapy. So, some have advocated that it is best to cool them off with chemotherapy first. We do not have definitive data confirming this, although when you look at the randomized data, there is a group of patients who seem to die earlier from the immunotherapy and do better with chemo­therapy. However, the survival curves then overlap, and you see the more durable and long-term benefits with an ICI. Many, myself included, would consider an ICI the standard of care.

The patient was treated with nivolumab (Opdivo) at 240 mg administered as an intravenous infusion over 60 minutes every 2 weeks.

Imaging at 3 and 6 months showed significant shrinkage at the pulmonary lesions. At 9 months, she complained of increasing fatigue and need for frequent rest.

Targeted Oncology: What type of monitoring is necessary for a patient receiving immunotherapy, and what are your impressions of this patient’s symptoms?

SIEFKER-RADTKE:The patient is responding well and having interval shrinkage at 3 to 6 months. With most of the ICIs, like nivolumab, you will see the responses in that early time interval. We do see evidence of benefit and often patients report improvement of symptoms as well.

Nine months later, she has increasing fatigue. When someone is presenting with fatigue, that would be a little bit unusual for a responding patient. I think it is important to consider the differential diagnosis. One possibility is that they could now be progressing. So, in a patient who is feeling more fatigued, I will often do CT imaging to make sure they are responding and it is not a symptom of progressive disease.

There can be other causes of fatigue as well, and you want to make sure that you do not miss adverse effects from an ICI. ICIs can cause thyroiditis. So, we do follow thyroid function, because if it is dropping they could be feeling more fatigued. Some patients initially get an increase in thyroid hormones, due to inflammation of the thyroid gland, which eventually burns out, resulting in hypo­thyroidism. We would typically give those patients with thyroid medication once their thyroid gland is no longer producing enough hormones for them. Other things on the differential include adrenal insufficiency. ICIs can attack multiple glands, including the thyroid and the adrenal gland, so you may want to consider an adreno­corticotropic hormone stimulation test to determine if they have evidence of adrenal insufficiency. If they do, you do not want to miss that diagnosis. You want to get the patient on replacement steroids right away. Other things on the differential include panhy­popituitarism; ICIs can cause an immune reaction against the pitu­itary gland, and you want to check for multiple hormones that are produced by the pituitary.

Fatigue is not just from attack of glandular organs, sometimes you can have an immune reaction against the nerves in which patients develop Guillain-Barré syndrome. Such a patient defi­nitely needs to be hospitalized and placed on steroids to keep the immune reaction turned down in order to keep them from developing full-blown paralysis requiring intubation. You should be looking for evidence that this might be impacting the nerves or causing a demyelinating condition, and an electromyography can be helpful to make that diagnosis.

A final thing to consider for someone who is fatigued is myositis. You can get inflammation against the muscle, so you may want to consider muscle biopsies. This includes a variety of muscle markers in the blood to determine if there is evidence of a myositis or general muscle inflammation. All of these have been seen and reported with ICIs.

After he had chemotherapy, about 10 months later, as many would anticipate since most chemotherapy responses are not durable, he developed progressive disease in the lung. His current liver function is normal. He does have moderate renal impairment with a CrCl of 49 mL per minute, which is common in patients with urothelial cancer since they have comorbid condi­tions often associated with the development of their cancer and decrements in renal function. Patients can be diabetic; they can have poor hearts and obstructed ureters; and many of them are smokers, which can result in a decline in renal function. This patient also has rheumatoid arthritis, which is currently managed with adalimumab and methotrexate. He is on active treatment with 2 agents for his rheumatoid arthritis. He has a pretty good ECOG performance status of 1.Targeted Oncology: What are the triggers to change therapy versus continuing on immunotherapy?

SIEFKER-RADTKE:If you do find evidence of an autoimmune condition, it depends on the type. For instance, if the patient develops thyroiditis, I haven’t seen any evidence yet that we can prevent burnout of the thyroid gland from happening. Likewise, with adrenal insufficiency or panhypopituitarism, when I see evidence of those, I typically do replacement hormones and continue the patient on the ICI.

However, if you see evidence of myositis, autoimmune conditions against the muscle, or Guillain-Barré, those are the types of reactions where I have typically stopped the ICI out of concern that continuing this could stimulate more of that auto­immune reaction and result in the patient ending up with long-term hospitalization or intubation due to difficulty breathing.

Targeted Oncology: Would you consider the 480-mg dose of nivolumab every 4 weeks?

SIEFKER-RADTKE:I personally would. Some physicians want to watch patients more closely, and some might argue, if you thought this was not myositis or Guillain-Barré, continue the every-2-week dose. But personally, if it was thyroiditis, I would consider giving them the 480-mg dose based on this dose’s recent FDA approval. The dose seems to maintain adequate serum levels for a long period of time. It also results in a conve­nience for the patient because they do not have to come in as frequently.


Durvalumab (Imfinzi). FDA website. ApprovedDrugs/ucm55930.htm. Updated May 1, 2017. Accessed April 5, 2018.