Subtypes of Renal Cell Cancer

The Pathology of Renal Cell Cancer

There are several sub-types of renal cell cancer and the prognosis and treatment can depend on what sub-type you have. There are also several modifying characteristics which, while not defining a separate subtype, can affect prognosis and treatment. The sub-type and any modifying characteristics are determined by a pathologist and aren’t known until a sample of your tumor is available for the pathologist to examine. This happens when there is a nephrectomy, surgical removal of metastatic tumors, or when a biopsy is taken. If you’ve just been diagnosed and haven’t had your kidney out, or had a biopsy, the pathologic sub-type will not yet be known. (For general information on pathology see Dr. Ed Uthman’s “The Biopsy Report” on CancerGuide.).

Although this is one of the most complex and detailed of my kidney cancer articles, don’t be intimidated! You only have to read the parts that apply to your sub-type.

Goals of This Article

  • Most articles on the pathology of RCC go into highly technical descriptions of the appearance of the various sub-types. I am avoiding detailed descriptions of this nature. I doubt you would find descriptions like “Psammoma bodies are commonly seen…” useful since you will presumably not be trying to “read” your own pathology slides and since, frankly, much of the specialized vocabulary of pathology is beyond me anyway.
  • Instead, my goals are to help you:
    • Sort through your pathology report to figure out what sub-type and other important characteristics you had.
    • Describe what is known about treatment and prognosis of specific sub-types, though often knowledge is very limited because some of these sub-types are so rare.
    • Determine if you need a pathology second opinion.

Steps to Take

  • Obtain a copy of your surgical pathology or biopsy report from your doctor. (If you’ve just been diagnosed it will be available after surgery or biopsy).
  • By all means, discuss your pathology report and your understanding of it with your doctors!
  • Use this article to help you figure out what sub-type and other characteristics you have and what this might mean for treatment.
  • Use this article to help determine if a pathology second opinion might be warranted. A second opinion might be in order if you have a rare sub-type, or if the classification of your tumor doesn’t use the most recent classification scheme, or simply wasn’t classified, or if it seems uncertain in some way. Any current pathology report should include both the subtype and Fuhrman Grade, although a surprising number from non-experts don’t [Carloss 2004].

Things to Consider

  • About 80% of renal cell cancers are clear cell. This is by far the most likely possibility, and most of the results of treatment of kidney cancer largely reflect results in clear cell RCC. All of the rest of my information on treatment applies if you have clear cell RCC. If you have something other than clear cell, study this page carefully and consider my other kidney cancer articles in the light of what you learn here.
  • If the disease is limited to the kidney, treatment is limited to surgery, or rarely surgery plus an adjuvant clinical trial. As with clear cell RCC, there is no standard treatment after the surgery to prevent recurrence. If you are have a rare sub-type and are also looking at adjuvant trials, it’s important to keep in mind whatever is known about how similar the response of your sub-type is to clear cell. In some cases, the appropriate follow-up schedule should be modified depending on the sub-type. If you have a rare sub- type, knowing that will be definitely important if it recurs.
  • Evidence on how to treat metastatic disease in rare sub-types, where it exists at all, is always much thinner than for clear cell (where it is thin enough already!), but you have to make the best of what evidence there is, and often there is enough to be useful.

A Good Free Review Article

Before plunging in, you might want to read The Mainz Classification of Renal Cell Tumors[Diaz 1999]. This review article from the technical literature is good background. It’s relatively easy reading (which doesn’t mean easy!) for a pathology article, and while it doesn’t say much about treatment implications it, describes the classification well.


General Advice on Treatment for Rare Sub-Types

Specific Subtypes

Clear Cell (About 83% of renal cell cancers [Cheville 2003])

This is “garden variety” RCC – by far the must common sub-type (though it’s anything but routine if you have it). The rest of my kidney cancer section covers treatment for RCC in general and all applies to you if you have clear cell RCC (with a few possible exceptions covered here).

Telling if it’s Clear Cell

  • Usually it’s clear from the pathology report that the diagnosis is clear cell, but there are some cases where it can be confusing. It’s important not to mistake clear cell for something else so if it’s not “clear”, discuss it with your doctors, and if necessary consider a pathology second opinion.
  • If you are said to have papillary and clear cell in some combination you may well have clear cell. The mere mention of the term “tubulo-papillary” or the mention of areas of papillary architecture should not be taken to mean you have papillary. Again discuss it with your doctors.
  • Granular and clear cell is still clear cell. Mixed clear and granular cell is very common.

Modifying Characteristics

There are two modifying characteristics that can be important in treatment and follow up, sarcomatoid and granular. If these are mentioned in your report be sure to check the appropriate section of this page.

Papillary (Chromophil) (About 11% of renal cell cancers[Cheville 2003])

Papillary renal cell carcinoma is the second most common sub-type after clear cell. The mere mention of “papillary” or “tubulo-papillary” in your pathology report does not necessarily mean you have papillary RCC. Papillary RCC is diagnosed based on the growth pattern of the cells in the tumor along with various other technical criteria. Clear cell RCC can have areas which have a papillary growth form but this doesn’t predominate in clear cell tumors. If you have true papillary RCC this should be clearly stated as the final diagnosis. Unfortunately, according to an informal comment by a well known kidney cancer expert, more than half of the slides sent to his institution with a diagnosis of papillary RCC turn out to be clear cell RCC with papillary features. Based on this, I think it’s appropriate for anyone who has metastatic disease and a diagnosis of papillary RCC to be sure their pathology has been read by an expert in the pathology of renal tumors. If you don’t have metastatic disease, then knowing for sure if it’s clear cell or papillary won’t change much since nephrectomy is the only standard treatment and any reputable clinical trial of an adjuvant therapy will include a review of your slides anyway.

Papillary RCC actually has two sub-types, type I and type II. This distinction is relatively recent [Delahunt 1997] , but recent pathology reports should specify type I or type II. Type I is also called “basophilic” and type II is also called “eosinophilic”. These names reflect standard dyes that are used to stain the tissue as part of the pathology exam. Eosinophilic means it takes up the dye Eosin. Basophilic means it takes up the dye hematoxylin which is a base. Eosin is acidic.

Type I is on average diagnosed at a lower stage, and lower grade than type II and also has a better prognosis. Type I appears to be more common than type II in general but in a study of patients being treated for metastatic RCC, Upton 2003 found 14 type II patients and only 2 type I patients. This remarkable reversal of the prevalence of type I and type II strongly suggests that type II metastasizes much more commonly than type I.

Papillary RCC seems overall to have a better prognosis than clear cell of the same stage and grade for lower stages and grades but about the same for high stage and grade which seem to do about as well as clear cell[Cheville 2003]. Given the enrichment in high stage and grade cases in type II, I suspect this actually is a reflection of a prognosis difference between type I and type II. In short, I think type I has a significantly better prognosis than clear cell RCC, and type II has about the same prognosis as clear cell.

Other than surgery, there is almost no information on the treatment of papillary RCC. This is surprising considering that papillary is the second most common sub-type. It may partially reflect a smaller than expected number of metastatic cases due to the better prognosis of most papillary RCC.

Localized Papillary RCC

Follow-Up Implications: Since papillary RCC seems to generate multiple primary kidney tumors more often than clear cell RCC[Levin 2000], I suggest that follow-up include appropriate imaging of the kidney(s).

IL-2 for Metastatic Disease

Unfortunately, IL-2 is unlikely to be a good option for patients with metastatic papillary RCC, at least for those who have type II papillary RCC. On the KIDNEY-ONC E-Mail List we had been hearing that most of the experts think papillary RCC doesn’t respond to IL-2 for some time, but were unable to find data in the literature to support this. At ASCO 2003, the first real data was presented. Upton 2003 reviewed response of the various sub-types of RCC to IL-2 based therapy. There were 14 patients with type II papillary RCC and 2 with type I. Unfortunately, none responded.

14 consecutive type II non-responders to IL-2 is enough to suggest that IL-2 works poorly in this sub-type. If there were a 20% response rate, only 5% of the time would you’d see no responses in 14 randomly selected patients. On the other hand, I don’t think no response in the two type I patients says much about the chance of response with type I. One weakness here is that the study included patients treated with a variety of IL-2 regimens and we don’t know how many of these got high dose IL-2, which seems to be the most potent form.

I do know of a single patient with type II papillary RCC who did respond to IL-2. This member of the KIDNEY-ONC E-Mail List had multiple tumors in both kidneys which shrank more than 50% after treatment with low dose bolus IV regiment (72,000IU/kg/dose, 10% of high dose). His response lasted 15 months before he relapsed. His pathology had been reviewed at NCI, so I believe the chance of a misdiagnosis is low.

Other Drug Therapies for Metastatic Disease

Unfortunately, even although papillary is the second most common sub-type of RCC, I have found no published data on other drug treatment just for papillary cell RCC.

Surgery for Metastatic Disease

Papillary RCC is often relatively slow growing, and especially given the lack of other options, I think aggressive surgery to remove metastatic disease is well worth considering, if all of it can be removed. Even if papillary RCC is biologically a different disease than clear cell RCC, surgery for metastasis is well known to be an effective treatment for such disparate forms of cancer as osteosarcoma and colon cancer when all disease can be removed. See our reviews of surgery for metastatic disease for results in metastatic RCC. Note that the data here unfortunately does not break out the results by sub-type.

Anti-Angiogenic Therapies for Metastatic Disease

Unlike clear cell RCC, Papillary RCC isn’t very vascular and it lacks the VHL mutation which causes the extreme vascularity of clear cell RCC. While this doesn’t automatically prove antiangiogenic drugs (and especially those targeting the VEGF pathway which is activated in clear cell RCC) won’t work, they are less of a natural fit for papillary RCC than for clear cell.

Chromophobe (About 4% of renal cell cancers [Cheville 2003])

Chromophobe RCC seems to have a better prognosis than clear cell for the same stage.

Follow-Up Implications: Chromophobe RCC tends to metastasize to the liver more than clear cell RCC [Levin 2000], so I suggest that follow-up include appropriate imaging of the liver.

I have found essentially no useful information on whether chromophobe RCC responds differently than clear cell RCC to immunotherapy or any other treatment. Given this, if possible, I would tilt more towards aggressive surgery when all metastases can be completely removed. I think surgery should be less likely to depend on details like the cell-type than immunotherapy. Since chromophobe tends to spread to the liver, I would consider special techniques for treating liver tumors, like radio frequency ablation or cryosurgery, if necessary. If surgery is not possible, then lacking other knowledge I think the best bet is still IL-2 immunotherapy, preferably high dose, but if at all possible, I would certainly consult a world expert and question closely about the rationale for any treatment for this rare form of kidney cancer.

Collecting Duct Carcinoma (Bellini Duct Tumor) (Less than 1% of renal cell cancers [Cheville 2003])

This is a very rare and aggressive subtype. It is so rare that there is almost nothing in the way of useful information on treatment. Most of the literature is in the form of individual case reports or very small case series.

Several papers say that it can be difficult to distinguish CDC from papillary RCC [Mejean 2003,Chao 2002], and patients who have cancer which is behaving differently from the diagnosis e.g. fast growing aggressive papillary RCC or slow growing indolent CDC may want to get a second pathology opinion.

Various case studies indicate that if collecting duct carcinoma (CDC) is caught before there are either lymph node or distant metastases, surgery to remove the primary tumor can bring long term survival and likely even a cure. For example, Mejean 2003 had three patients with apparently localized CDC, and two of the three were long term disease free survivors at 99 and 100 months post surgery.

Metastatic CDC has a poor prognosis. It’s so rare though that there is very little information on what’s been tried so it’s hard to know what works. The literature on this sub-type is so thin that it’s also hard know what won’t work – and so really there’s a hope that almost anything might, especially in an individual case.


There is a little evidence that immunotherapy can work in CDC.Dimopoulos 1993 reported a single complete response to outpatient IL-2 + Interferon which was complete and ongoing at 30 months out of six CDC patients treated with this regimen. Upton 2003reported a single response to IL-2 based immunotherapy in a patient with CDC. We don’t know which IL-2 based therapy was used, nor whether the response was partial or complete, nor whether it was lasting. Overall there is a suggestion IL-2 based immunotherapy can be effective in CDC. Given that IL-2 treatment is well documented to produce long term durable response in RCC in general I think this “suggestion” should be given considerable weight.

Cytotoxic Chemotherapy

Since CDC is a fast growing tumor, it might be susceptible to cytotoxic chemotherapy. There is a single report in the literature of metastatic CDC responding to chemotherapy with adriamycin + gemcitibine[Milowsky 2002] but unfortunately the response only lasted a few months. With only a single patient, we don’t know if the possibilities for this regimen might include longer responses as has been the case for sarcomatoid RCC [Nanus 2003]. More encouragingly, there is a case report of “long term” remission in a young CDC patient treated with the cytotoxic chemotherapy drugs taxol and carboplatin, followed by surgery to remove residual disease in the kidney[Gollob 2001]. This patient was disease free at 20 months. Another study Chao 2002reported two patients who were treated with the same drugs. One had a partial response which lasted 6 months while the other had no benefit.

Since so little is known about the responsiveness of CDC to chemotherapy, it might make sense to use chemosensitivity testing to determine which cytotoxics the tumor might be responsive to and which it might be resistant to. Chemosensitivity testing is controversial, but I think it makes the most sense where little is known about what might work, as is the case here. One major disadvantage is that it requires a fresh sample of the tumor which will be easier to get in some cases than in others.

For more on Chemosensitivity Testing from a major proponent, see Larry Weisenthal’s Essay on Chemosensitivity Testing

Surgery for Metastatic Disease

Because this disease tends to be so fast growing and aggressive, I think surgery for metastatic disease is less likely to be effective and suggest other therapy first. If there has been a major response to drug therapy and all remaining disease can be removed, that makes sense, and was part of the treatment of the patient with long term response to taxol/carboplatin I referred to above.

Rare Variants

Believe it or not, even this very rare sub-type has its own even more unusual variants. Two other closely related cancers which are either sub-types of CDC or similar to CDC are Renal Medullary Carcinoma and Low Grade Collecting Duct cancer.

  • Renal Medullary Carcinoma is an extraordinarily rare cancer which occurs in black people with the sickle cell trait. Patients tend to be young. This tumor has an even more aggressive behavior and worse prognosis than regular CDC [Mejean 2003,Levin 2000].
  • Low Grade Collecting Duct Cancer is just what it sounds like and seems to have a better prognosis than CDC [Mejean 2003].

Renal Oncocytoma

Although Renal Oncocytoma sounds especially sinister, this is actually a benign tumor! It is not a cancer at all. On the KIDNEY-ONC E-Mail List we have encountered patients who didn’t realize they actually didn’t have cancer.

Renal Oncocytoma occurs at about 5 to 10% of the rate of kidney cancer.

Renal oncocytoma can occasionally be confused with chromophobe RCC or the granular variant of clear cell RCC, although most cases are easily diagnosed.

Although benign, oncocytoma occasionally co-exists with cancer which may be present within or near the oncocytoma (and which I think could be missed). For this reason, I think patients with oncocytoma should be followed even though the tumor is benign, and the prognosis is very, very good.

My information on this tumor is from Levin 2000.

Modifying Characteristics


“Sarcomatoid” means cells that look much like the cells of sarcomas. You may also see sarcomatoid cells referred to as “spindle cells”. Sarcoma is a broad class of tumors which includes tumors of muscle, bone, and connective tissue, but the literature differs on whether sarcomatoid cells are RCC cells which merely resemble sarcoma, or whether they are actually derived from the tissues which give rise to sarcomas [Ro 1987]. I think it’d be very strange for two tumors to regularly arise in the same organ at the same time, and think it’s more likely that sarcomatoid cells merely resemble sarcoma (See my comment on Ro 1987). Interestingly though, sarcomatoid RCC can respond to chemotherapy, and most of the chemo regimens which have worked with sarcomatoid RCC are primarily used to treat sarcoma.

Sarcomatoid RCC is an aggressive and fast growing tumor, but treatment can sometimes be very effective and there is real reason for hope with sarcomatoid RCC. There is also a bit more information on treating sarcomatoid RCC than some of the other sub-types, though still not nearly enough.

Sarcomatoid cells almost always make up only a part of the primary tumor. The rest of the tumor can be of any of several sub-types, most commonly clear cell, but papillary and chromophobe are also possible. Thus sarcomatoid modifies the sub-type, but is not a distinct sub-type on its own. A diagnosis of sarcomatoid tumor in the kidney tumor should be coupled with a description of the main sub-type. If it is not, or if it is claimed that the primary tumor is purely sarcomatoid, a more careful examination of more sections of the tumor may reveal the main sub-type [Levin 2000]. Metastases can be purely sarcomatoid.

The proportion of sarcomatoid tumor may be important in determining the best treatment, and if there is sarcomatoid tumor some indication of the proportion should be reported.

Treatment of Localized Sarcomatoid RCC

Sarcomatoid RCC clearly has a worse prognosis than clear cell even when apparently localized, and patients with large (stage III) tumors all (or almost all) relapse within a few years. [Ro 1987,Mian 2002] Patients with stage I tumors, do better [Ro 1987]and while the risk is still high, it may be that some are cured by surgery. Patients with a smaller proportion of sarcomatoid tumor also do better. If you are “lucky” enough to have a small tumor with a low proportion of sarcomatoid RCC I would guess you may be cured by surgery although there isn’t enough data to absolutely prove it (A large tumor with a small sarcomatoid proportion is rare and it’s hard to guess if surgery offers a chance of cure – the risk of relapse would be high because it’s high in stage III RCC even without sarcomatoid involvement). Almost all other patients with sarcomatoid RCC will relapse. In short if you have sarcomatoid RCC you need to be prepared for the likelihood of relapse.

Follow-Up Implications: Because of the especially high risk of recurrence I think follow-up should be more frequent and more intensive than usual.

Adjuvant Therapy: There is no proven form of therapy to prevent recurrence for any form of RCC and certainly not sarcomatoid RCC, but in several other cancers where chemo works in metastatic disease there is even more benefit when it is used as adjuvant therapy. So given the extreme risk of recurrence, combined with the data showing some cases of metastatic sarcomatoid RCC do respond well to cytotoxic chemotherapy, I wonder if adjuvant cytotoxic chemotherapy should be considered for stage II and III cases. I don’t feel qualified to actually recommend this but suggest you consult with a specialist who has experience with sarcomatoid RCC.

Treatment of Metastatic Sarcomatoid RCC

IL-2 Based Immunotherapy

Several studies have shown that sarcomatoid RCC can respond to IL-2 based immunotherapy [Cangiano 1999Upton 2003Mian 2002]. Data on the quality and duration of responses are hard to come by in part due to poor reporting, but I see no reason a priori to assume the responses would be less durable is seen in RCC in general. One paper hints that responses to high dose IL-2 can be relatively durable [Cangiano 1999]while another suggests responses to outpatient combinations of IL-2 with cytotoxic drugs are rarely durable [Mian 2002]. Therefore, I think high dose IL-2 would be the best shot.

Cytotoxic Chemotherapy

Cytotoxic chemotherapy has been notoriously ineffective in kidney cancer, but sarcomatoid RCC can be a definite exception. There have been complete responses to chemo in sarcomatoid RCC, some of which have been ongoing after years and could well be cures. All of the long-term responses I am aware of were to chemo regimens which included the potent cytotoxic drug, doxorubicin (adriamycin) [Bangalore 2001,Culine 1995Lupera 1989,Nanus 2003Sella 1987]. It’s hard to know what the chances of such a response are since these are mostly individual case reports or very preliminary results of small trials.

One of these small trials, Escudier 2002, had no responses at all. It combined doxorubicin and ifosphamide. Some of the long term responses were to a combination of these drugs with two others (the MAID protocol, which also includes methotrexate and dacarbazine). It’s not clear whether this poor result is due to bad luck, a small chance of response, selection of patients with low sarcomatoid proportion (see below), or simply that this doxorubicin containing regimen does not work. In any event, if you choose chemo, choose one of the regimens which has resulted in long term remission (see the references cited above).

Chemotherapy or IL-2?

A study presented at the 2003 ASCO meeting [Upton 2003]showed that patients who have a high proportion of sarcomatoid cells in their kidney tumor had little chance of responding to Il-2 based immunotherapy, while it appeared those with a lower proportion of sarcomatoid cells had as good a chance to respond to IL-2 as anyone else. At the same time, there have been several reports lasting complete response to cytotoxic chemotherapy in sarcomatoid RCC even though ordinary RCC is notoriously resistant to cytotoxic chemo.

I presume that sarcomatoid cells are resistant to immunotherapy, but potentially sensitive to chemotherapy, and that non sarcomatoid clear cells are resistant to chemotherapy, but potentially sensitive to immunotherapy. Since it is the metastases which are being treated presumably what matters is whether the metastases are sarcomatoid or not. I interpret the Upton study to mean that when the proportion of sarcomatoid cells is high in the kidney tumor the metastases are very likely to be sarcomatoid, while non-sarcomatoid metastases are likely if the sarcomatoid proportion is low in the kidney tumor.

It seems clear that if there is no other information, if you had a low proportion of sarcomatoid in their primary tumor you should choose IL-2, and if you had a high proportion you should choose chemotherapy. If you have information on whether the metastases are sarcomatoid from a biopsy or surgical specimen you may want to choose based on this evidence. I would consult carefully with your doctor on the accuracy of needle biopsy procedures which remove only a small amount of tissue.

If you have non-clear cell sarcomatoid, then I recommend choosing as above, unless there is evidence the other sub-type doesn’t respond to IL-2 in which case chemo makes more sense.

Surgery for Metastatic Disease

Surgery can be one of the most effective treatments for metastatic RCC (See our reviews at Surgery for Metastatic RCC), but I haven’t been able to find much useful data specific to sarcomatoid RCC. The largest experience I’ve found is the large case series of Mian 2002which included 21 patients who had surgery for metastatic disease. Unfortunately, they say almost nothing about either the results or the indications for surgery. They do say that two patients who had drug treatment were disease free after “resection of lung metastasis” and it sounds like they may have been long term survivors, perhaps even the only ones. It’s unclear whether these patients had surgery for residual tumor after a partial response or whether surgery was used after drug treatment failed. These 21 patients made up 19% of the case series, so it must at least be that surgery is possible in a significant fraction of those with metastatic sarcomatoid RCC. They did say 3 of the 21 patients had more than one surgery – implying that relapse after surgery can be limited enough to make another surgery reasonable. This is frequently the case for surgery in metastatic RCC in general.

Culine 1995 did report disease free survival after surgery for a solitary lung metastasis – I am in the process of getting the details.

With almost no data my best guesses on surgery are:

  • Given the very high percentage of patients with apparently localized who relapse after nephrectomy, it seems inescapable that most patients who have metastasis at diagnosis have truly wide spread disease and won’t benefit from surgery to remove the metastasis. The same would likely be true for patients who have a recurrence shortly after nephrectomy.
  • If it has been several years between nephrectomy and recurrence, this is evidence of a slower growing tumor which behaves more like non-sarcomatoid RCC and surgery would seem to make sense if the recurrence is limited, and all of the metastases can be removed.
  • Surgery for limited metastasis remaining after drug treatment seems reasonable, especially if there was a partial response or a significant period of stability.


Clear cell RCC often has areas of “granular” cells mixed in. If your pathology report says you have mixed granular and clear cell RCC, then the main sub-type is still clear cell. If the final pathologic diagnosis is granular cell RCC and there is no mention of clear cell then you may actually have another sub-type [Levin 2000]. In this case, a second pathology opinion is definitely in order.

Mixed clear and granular cells should have little impact on follow-up or prognosis in clear cell RCC limited to the kidney, and for most patients should not affect the treatment of metastatic disease either.

Mixed granular cell and clear cell is actually quite common. Out of 146 patients with metastatic clear cell tumors treated with IL-2,Upton 2003 found 39% had no granular cells, 47% had some granular cells but less than 50%, and 13% had more than 50% granular cells.

Metastatic Disease

If the proportion of granular cells in the kidney tumor was high, IL-2 may be much less likely to work. Upton 2003 recorded only one response to IL-2 based treatment in the 19 patients with more than 50% granular cells. Some granular cells, but less than 50%, only modestly affected the response rate, and in my opinion are no reason to avoid IL-2. We don’t know how many of the patients with a high proportion of granular cells in this study got high dose IL-2 (probably the most potent IL-2 regimen) and we also don’t know that 50% is the best cut-off. As above, most patients with mixed granular and clear cell have less than 50%, and at this time there is no evidence to show that this is the best cut-off.

Given the lack of good treatment options for renal cell cancer, and these uncertainties, and the fact that mixed granular cell tumors can respond, I don’t recommend avoiding IL-2 due to the presence of granular cells, unless the percentage of granular cells is especially high (I wouldn’t assume 50% is the best cut-off). I had mixed granular and clear cells (percentage unknown) and responded wonderfully. If granular cell is mentioned in your pathology report, but no indication of the proportion of granular cells is given, it would make sense to have it looked at again with an eye to determining this.


References Relevant to Multiple Sub-Types

Carloss E, Regan M, Upton M, Youmans A, McDermott D, Polivy A, Atkins M.
Concordance of local pathology interpretation (LR) with central pathology review (CR) in primary tumor specimens from patients with renal cancer (RCC).
Proc Am Soc Clin Oncol (ASCO Meeting Abstract) 2004 23 : Abstract 4548 [You can find this abstract on the ASCO website (I can’t link directly because ASCO won’t provide a stable URL)]

Comment: This study compared pathology reports from various hospitals in the community to reports on the same kidney tumors by one expert at a major center. Most of the cases were since 1994. A surprising number of the local reports didn't mention the subtype or give the Fuhrman grade at all. When the Fuhrman grade was mentioned, the expert assigned a higher grade in nearly half the cases. There was also a significant percentage of disagreement over whether there was sarcomatoid. One reassuring statistic: When the local report said it was clear cell, the expert agreed 99% of the time. I don't think there were enough papillary cases to tell if what we've heard informally about over-identification of papillary by local pathologists is true or not.

In theory it's impossible to know who is "right" when two reports disagree even though one is from an expert. This might be more certain if several experts had evaluated the specimens. Although simply missing something like grade or subtype is beyond dispute, since standards have likely improved over time (for instance as recognition of the importance of these factors has increased) this report might overstate the current odds of getting a deficient pathology report from your local hospital. Still, at least it's a strong motivation to make sure grade and sub-type are noted on your report and perhaps to have your pathology re-read by an expert if you are said to have other than clear cell, and are about to make treatment decisions based on that.

Cheville JC, Lohse CM, Zincke H, Weaver AL, Blute ML.
Comparisons of outcome and prognostic features among histologic subtypes of renal cell carcinoma.
Am J Surg Pathol. 2003 May;27(5):612-24. [PubMed Abstract (will open in new window)]

Comment: This huge study of the outcomes for RCC subtypes is "profusely illustrated" with survival curves showing survival by sub-type separated by stage and grade and is the best thing I've found on prognosis for the sub-types in non-metastatic disease.

I also used it for estimates of the prevalence of each sub-type. Since this study is limited to nephrectomy specimens, it leaves out those who present with metastatic disease and thus is likely to underestimate the prevalence of more aggressive sub-types which are often metastatic at diagnosis and likewise overestimates the prevalence of less aggressive sub-types. Although many of the patients were treated long ago, again most did not have metastatic disease and even the best current treatments would have little impact on this kind of survival curve so I think this is roughly valid for prognosis of non-metastatic disease in the various sub-types.

Diaz JI, Mora LB, Hakam A.
The Mainz Classification of Renal Cell Tumors
Cancer Control. 1999 Nov;6(6):571-579. [PubMed Abstract (will open in new window)]

Comment: This updated review article on the pathology of renal cell cancer is more readable than [Thoenes 1986] and full text is available on-line (see the reference above for the link).

Levin HS, Myles JL.
The Pathology of Renal Neoplasms
Ch. 2 in Renal Cell Carcinoma, Molecular Biology, Immunology and Clinical Management, Bukowski RM, Novick AC Eds 2000 ,Humana Press IBSN 0-89603-781-9.

Comment: A reasonably readable introduction to the sub-types, though difficult to access since it's a book chapter rather than a paper.

Thoenes W, Storkel S, Rumpelt HJ.
Histopathology and classification of renal cell tumors (adenomas, oncocytomas and carcinomas). The basic cytological and histopathological elements and their use for diagnostics.
Pathol Res Pract. 1986 May;181(2):125-43. [PubMed Abstract (will open in new window)]

Comment: This study is the basis of modern classification of kidney tumors. It is the original (English) article defining what is now known as the "Mainz Classification" of renal cell cancer. It is long and extremely technical.

Upton MP, Parker RA, Youmans A, Connolly C, Atkins MA
Histologic predictors of renal cell carcinoma (RCC) response to interleukin-2-based therapy
Proc Am Soc Clin Oncol (ASCO Meeting Abstract) 2003 22 : Abstract 3420 [You can find this abstract on the ASCO website (I can’t link directly because ASCO won’t provide a stable URL)]

Comment: This is the only significant study of how different sub-types respond to IL-2 based treatment and is therefore of great importance. The abstract referenced above has little of the essential information contained in the meeting poster. My reference to this study usually refers to data in the poster rather than the abstract.

Some key findings:

  • There were no responses in 14 cases of type II papillary RCC, strongly suggesting IL-2 is not effective in papillary RCC.
  • Also there were no responses in 9 patients with clear cell RCC with "papillary features". In my opinion this isn't enough to prove patients with this characteristic don't respond, but it suggests they may not.
  • There were 7/23 responses in patients with sarcomatoid/clear cell RCC who had less than 50% sarcomatoid involvement of their kidney tumor, but only one response in 15 with more than 50% involvement.
  • In patients with clear cell RCC who had granular cells in their kidney tumor, there were 13/67 responses in patients with less than 50% granular cells, while only 1 in 19 patients with more than 50% granular cells responded.
Despite the significance of this study, some important details are missing since it was only a meeting poster, and hasn't yet been published as a full paper. These include lack of information on which IL-2 based regimens these patients got, whether the responses were complete or partial, and how durable they were.

Papillary RCC References

Delahunt B, Eble JN.
Papillary renal cell carcinoma: a clinicopathologic and immunohistochemical study of 105 tumors.
Mod Pathol. 1997 Jun;10(6):537-44. [PubMed Abstract (will open in new window)]

Comment: This is the original paper reporting the discovery that papillary RCC can be divided into type I and type II. Type I tends to be diagnosed at a lower grade and stage compared to type II, suggesting it has a better prognosis. Happily, Type I appears to be more common than type II. The paper goes into great detail about how to tell the two types apart, but there is no information on treatment.

Delahunt B, Eble JN, McCredie MR, Bethwaite PB, Stewart JH, Bilous AM.
Morphologic typing of papillary renal cell carcinoma: comparison of growth kinetics and patient survival in 66 cases.
Hum Pathol. 2001 Jun;32(6):590-5. [PubMed Abstract (will open in new window)]

Comment: This paper by the doctors who discovered that papillary RCC has two sub-types [Delahunt 1997] confirms that the prognosis is better for type I than type II. Additionally, and not surprisingly, they showed that type II has a higher fraction of dividing cells and thus grows faster. Again, there isn't any information on treatment here.

Sarcomatoid RCC References

Bangalore N, Bhargava P, Hawkins MJ, Bhargava P.
Sustained response of sarcomatoid renal-cell carcinoma to MAID chemotherapy: case report and review of the literature.
Ann Oncol. 2001 Feb;12(2):271-4. [PubMed Abstract (will open in new window)]

Comment: This patient's primary tumor was more than 95% sarcomatoid and needle biopsy of a metastasis showed sarcomatoid RCC. He had rapid progression of his disease with high dose IL-2, but had a complete response to MAID chemotherapy, and was in complete remission over four years later. This is a fine example of high proportion sarcomatoid RCC responding wonderfully to chemotherapy and not at all to immunotherapy.

Cangiano T, Liao J, Naitoh J, Dorey F, Figlin R, Belldegrun A.
Sarcomatoid renal cell carcinoma: biologic behavior, prognosis, and response to combined surgical resection and immunotherapy.
J Clin Oncol. 1999 Feb;17(2):523-8. [PubMed Abstract (will open in new window)]

Comment: This important paper on sarcomatoid RCC out of UCLA reports on 31 cases which makes it one of the largest experiences reported. Unfortunately, it is confusingly written. What is clear is that sarcomatoid RCC can respond to IL-2, and that's important.

They had two partial responses out of 9 patients treated with high dose IL-2 alone, no response in 5 patients treated with low dose IL-2 alone, and one complete and two partial responses in patients treated with a combination of low dose IL-2, Interferon, and Tumor Infiltrating Lymphocytes (TIL). TIL is not commonly used at this time, and isn't easily available. There is poor information on the duration of these responses, but some of the partial responses were ongoing after a few years. The complete response was ongoing after at least 46 months.

The survival curve for the 9 patients treated with high dose IL-2 is spectacular - with 100% survival at 10 months and about 80% survival after 20 months (the end of the curve). The problem is that this can't be accounted for by the two partial responses in these 9 patients (I believe with IL-2 response translates to survival). They give no real explanation, and I'm skeptical. I think the short follow-up of a significant number of these patients combined with a little luck could account for this. Likewise, they claim that adjusted for other factors ,the relative risk of death was 10.4 times more in patients who didn't get high dose IL-2 than in the other patients, but they don't say how the data were "adjusted", or give a confidence interval or statistical significance for this provocative number. Many statistical tests and methods become quite squirrelly with small sample sizes, and again there is no explanation of the difference.

Bottom line: Though the numbers are very small, the response rate for high dose IL-2 here was consistent with the usual response rates for high dose IL-2, and moreover there was nothing to suggest these responses were particularly short lived.

Culine S, Bekradda M, Terrier-Lacombe MJ, Droz JP.
Treatment of sarcomatoid renal cell carcinoma: is there a role for chemotherapy?
Eur Urol. 1995 ;27(2):138-41. [PubMed Abstract (will open in new window)]

Comment: (Based on abstract only) Another case report including extended survivals in some patients treated with adriamycin based chemotherapy. They also had one patient who was disease free after surgery for a solitary lung metastasis - one of the few reports of resection of metastatic disease in sarcomatoid RCC.

Escudier B, Droz JP, Rolland F, Terrier-Lacombe MJ, Gravis G, Beuzeboc P, Chauvet B, Chevreau C, Eymard JC, Lesimple T, Merrouche Y, Oudard S, Priou F, Guillemare C, Gourgou S, Culine S
Doxorubicin and ifosfamide in patients with metastatic sarcomatoid renal cell carcinoma: a phase II study of the Genitourinary Group of the French Federation. of Cancer Centers
J Urol. 2002 Sep;168(3):959-61. [PubMed Abstract (will open in new window)]

Comment: Adriamycin (doxorubicin) and ifosphamide are the "AI" in the MAID chemotherapy regimen, which has been responsible for some long term survivals with sarcomatoid RCC. This is certainly the largest prospective trial just for sarcomatoid RCC with 25 patients - and maybe the only one. Most of these patients had more than 50% sarcomatoid RCC where I would expect chemo to be most likely to work, but unfortunately, this trial was a complete failure with not even one response. My best guess is that this particular chemo just doesn't work for sarcomatoid RCC, even though it does include doxorubicin. It must be that the Methotrexate and Dacabazine are necessary for MAID to work.

Lupera H, Theodore C, Ghosn M, Court BH, Wibault P, Droz JP.
Phase II trial of combination chemotherapy with dacarbazine, cyclophosphamide, cisplatin, doxorubicin, and vindesine (DECAV) in advanced renal cell cancer.
Urology. 1989 Nov;34(5):281-3. [PubMed Abstract (will open in new window)]

Comment: (based on abstract only) This report of a small clinical trial for metastatic RCC in general recorded a single complete response in a patient with sarcomatoid RCC. Because this trial was not specific to sarcomatoid RCC this basically amounts to a case report.

Mian BM, Bhadkamkar N, Slaton JW, Pisters PW, Daliani D, Swanson DA, Pisters LL.
Prognostic factors and survival of patients with sarcomatoid renal cell carcinoma.
J Urol. 2002 Jan;167(1):65-70. [PubMed Abstract (will open in new window)]

Comment: With 108 patients, this paper from MD Anderson cancer center is the largest case series for sarcomatoid RCC I know of. Patients were treated with a vast array of different therapies. Most were treated with nephrectomy and some form of immunotherapy usually combined with chemotherapy. In fact, 18 different forms of immunotherapy were used! This included IL-2, usually combined with other drugs such as interferon, and various chemo drugs, as well as Interferon also usually with other chemo drugs. High dose IL-2 was not used. They had very reasonable response rates of about 1 in 3 patients with either IL-2 or Interferon based therapy, but all the responses were partial. Frustratingly, they present no direct data on response duration.

Reading between the lines, the relatively high response proportion combined with survival curves showing a precipitous initial decline, as well as the fact that all responses were partial, strongly suggests most responses were not durable. Since sarcomatoid RCC is known to respond to chemotherapy, these could mainly represent brief responses to the chemo part of the chemoimmunotherapy treatments must patients received. The chemo drugs used were drugs like 5-FU which have often been combined with IL-2 rather than the adriamycin based regimens which have accounted for a few durable response in sarcomatoid RCC.

Although survival was very poor, they did have a few long term survivors even with metastatic disease. Their survival curves run out to 96 months and show that, despite the poor odds, there is a long term chance. Also these curves flatten out after 1 to 3 years. What is extremely frustrating is that they do not describe the characteristics of the long term survivors or how these patients were treated. They also state they resected metastatic disease in some patients but then don't say anything useful about the results ,except for a very few patients who had surgery for residual disease after immunotherapy and did relatively well. In fact, these may have been the only long term disease free survivors among their patients with metastatic disease.

The paper also included patients who were treated for localized sarcomatoid RCC. Of the 108 patients, 25 had tumors apparently localized to the kidney. There was a high proportion of T3 tumors (60%). Most of their patients relapsed and relatively quickly (median time to relapse 5.8 months). The three patients who didn't relapse had follow-up times of 22, 29, and 68 months. Again, they don't say much about the influence of disease stage or the amount of sarcomatoid tumor. They do suggest aggressive follow-up scanning in localized sarcomatoid RCC.

Nanus DM, Garino A, Milowsky MI, A. Rosmarin, Larkin M, Dutcher JP.
Active chemotherapy for aggressive renal cell carcinoma (RCC)
Proc Am Soc Clin Oncol (ASCO Meeting Abstract) 2003 22 : Abstract 1635 [You can find this abstract on the ASCO website (I can’t link directly because ASCO won’t provide a stable URL)]

Comment: This study from ASCO 2003 reported a relatively high response rate in patients with fast growing RCC, mostly sarcomatoid, treated with adriamycin and gemcitibine chemotherapy. One patient had a complete response which was ongoing at 21 months. This patient had locally recurrent sarcomatoid RCC (Personal communication from the patient's spouse), the other responses were partial. My impression was that follow-up on these partial responses tended to be short, and it is unclear how much benefit these partial responses will be in the end.

I spoke to Dr. Nanus at ASCO and he told me that he thought the Gemcitibine was a "major player" in these results, though unfortunately, I neglected to ask why. The complete response fits the pattern of occasional lasting responses to adriamycin based chemotherapy.

Ro JY, Ayala AG, Sella A, Samuels ML, Swanson DA.
Sarcomatoid renal cell carcinoma: clinicopathologic. A study of 42 cases.
Cancer. 1987 Feb 1;59(3):516-26. [PubMed Abstract (will open in new window)]

Comment: This study has useful information on the prognosis of localized sarcomatoid RCC. By stage, stage I patients had much better survival then stage III or IV, and stage II patients were intermediate. While their stage I patient with longest follow-up apparently relapsed, survival as long as 80 months was seen. Given the small number of stage I patients, this doesn't exclude cure for a few. Patients who had a lower proportion of sarcomatoid also tumor did better, and interestingly patients with minimal or no necrosis (dead areas) in their tumors did better than those with moderate or massive necrosis.

All of 13 their patients with stage III tumors relapsed within two years or so. Their 9 stage II patients also all relapsed, but not as fast.

This is the only paper I've found which describes the histology of metastases from sarcomatoid RCC. Of 12 cases, 4 had purely sarcomatoid metastasis, 3 had purely carcinoma (I presume mostly clear cell) and 5 had mixed carcinoma and sarcomatoid elements. This supports the idea that treatment for clear cell such as immunotherapy can work when the metastases are clear cell, and also that chemotherapy might work when they are sarcomatoid. The presence of mixed clear cell and sarcomatoid metastases is interesting. I think this means metastases which start from non-sarcomatoid (but genetically unstable) cells which require few additional changes to assume identifiable sarcomatoid characteristics can then evolve a new tumor with both elements (presumably sarcomatoid evolving to clear cell is not likely!). The other possibility is that the metastasis was started by more than one escaped cell, but according to the premier cancer text book (DeVita & Hellman, Cancer: Principles and Practice of Oncology), studies in a variety of cancers all show that metastases usually arise from a single cell.

Sella A, Logothetis CJ, Ro JY, Swanson DA, Samuels ML.
Sarcomatoid renal cell carcinoma. A treatable entity.
Cancer. 1987 Sep 15;60(6):1313-8. [PubMed Abstract (will open in new window)]

Comment: This study from MD Anderson reports on 44 patients who received various forms of treatment (oddly including 13 who got no treatment). It pre-dates the era of IL-2 immunotherapy. It also does not overlapMian 2002 (also out of MD Anderson).

The primary findings of interest are that 2 of 9 patients with metastatic sarcomatoid RCC who were treated with an adriamycin containing chemo regimen called CYVADIC were long term complete responders at 50 and 65 months (CYVADIC is cyclophosphamide, vincristine, adriamycin, and dacarbazine). Adriamycin seems to be the common theme in sarcomatoid patients who've gotten long term responses to chemotherapy.

Also of interest, although all four patients they treated with interferon died, they had longer survival than most of the others, living 25, 34, 48 and 50 months.

Finally, they had 19 patients with apparently localized sarcomatoid RCC and discouragingly all of them relapsed. Since they don't describe either the percentage of sarcomatoid involvement or the stage distribution, this doesn't prove that either early stage sarcomatoid tumors, or tumors with small amounts of sarcomatoid involvement have such a poor prognosis.

Chromophobe RCC References

C. L. Corless, M. C. Heinrich, S. Dimitrijevic, S. Dirnhofer, Z. Nikolova, J. A. Fletcher
Correlation of imatinib response with activation of KIT and PDGF receptors in a variety of cancers: Results of the CSTIB2225 trial
Proc Am Soc Clin Oncol (ASCO Meeting Abstract) 2003 22 : Abstract 783 [You can find this abstract on the ASCO website (I can’t link directly because ASCO won’t provide a stable URL)]

Comment: A fair number of the patients in this study on various types of cancer tested positive for KIT expression but still didn't respond (from the meeting poster). Apparently response to Gleevec is associated with not only expression of a targeted receptor but also an activating mutation (Conversation with Dr. Corless at the 2003 ASCO meeting). This suggests we should not assume that Gleevec would work in Chromophobe RCC even though it over expresses KIT. See also my comment for Yamazaki 2003.

Vuky J, Fotoohi M, Isacson C, Dela Cruz J, Otero H, Picozzi V, Aboulafia D, Malpass T, Jacobs A
Phase II trial of imatinib mesylate (formerly known as STI-571) in patients with metastatic renal cell carcinoma (RCC).
Proc Am Soc Clin Oncol (ASCO Meeting Abstract) 2003 22 : Abstract 1672 [You can find this abstract on the ASCO website (I can’t link directly because ASCO won’t provide a stable URL)]

Comment: The one KIT positive patient had chromophobe RCC but, like the other patients didn't respond, though was stable for 6 months.

Yamazaki K, Sakamoto M, Ohta T, Kanai Y, Ohki M, Hirohashi S.
Overexpression of KIT in chromophobe renal cell carcinoma.
Oncogene. 2003 Feb 13;22(6):847-52. [PubMed Abstract (will open in new window)]

Comment: This paper is misleadingly titled since it gives information on the gene expression profile of each sub-type of RCC and lists various genes for each sub-type which are overexpressed compared to normal kidney cells. This might be useful for determining targets for therapy. Nonetheless they did find strong overexpression of KIT in each of the four cases of chromophobe RCC they tested. They also found that, though overexpressed, KIT was not mutated in these tumors so we can predict a poor response to Gleevec (See my comment on Corless 2003).

Collecting Duct RCC References

Chao D, Zisman A, Pantuck AJ, Gitlitz BJ, Freedland SJ, Said JW, Figlin RA, Belldegrun AS.
Collecting duct renal cell carcinoma: clinical study of a rare tumor.
J Urol. 2002 Jan;167(1):71-4. [PubMed Abstract (will open in new window)]

Comment: Their only long term survivor had T1N0M0 grade 2 disease. All the rest of these patients had stage advanced disease at diagnosis. Two patients got taxol and carboplatin chemotherapy and one had a partial response which lasted six months. Combined with [Gollob 2001]this hints that chemo with taxol and carboplatin has activity in CDC. Unfortunately, unlike Gollob, this paper doesn't give the dose and schedule, so we don't know if the treatments were the same even though the drugs were.

Dimopoulos MA, Logothetis CJ, Markowitz A, Sella A, Amato R, Ro J.
Collecting duct carcinoma of the kidney.
Br J Urol. 1993 Apr;71(4):388-91. [PubMed Abstract (will open in new window)]

Comment: This study of 12 patients with CDC hints that immunotherapy can be effective. They treated 6 of their 12 patients with outpatient IL-2 + Interferon, and one had a nearly complete response with complete resolution of lymph node metastasis and had only 5% viable tumor in the diseased kidney which was removed after immunotherapy. This patient's complete response (after surgery) was ongoing at 30 months. Two of the other immunotherapy patients had stablizations lasting 10 and 15 months. One other patient treated with Interferon + 5-FU and Mitomycin-C had stable disease lasting 16 months. No significant benefit was seen for chemo only. They aren't very specific but say the most common regimen used was MVAC (Methotrexate, Vinblastine, Doxorubicin, Cisplatin)

This study otherwise confirms the aggressiveness of CDC. All four of their patients who had apparently localized disease relapsed within a year and 11 of their 12 patients died.

Gollob JA, Upton MP, DeWolf WC, Atkins MB.
Long-term remission in a patient with metastatic collecting duct carcinoma treated with taxol/carboplatin and surgery.
Urology. 2001 Dec;58(6):1058. [PubMed Abstract (will open in new window)]

Comment: This metastatic CDC patient had a simply spectacular result with chemo combined with surgery to remove residual disease!

Mejean A, Roupret M, Larousserie F, Hopirtean V, Thiounn N, Dufour B.
Is there a place for radical nephrectomy in the presence of metastatic collecting duct (Bellini) carcinoma?
J Urol. 2003 Apr;169(4):1287-90. [PubMed Abstract (will open in new window)]

Comment: I cite this mainly because it does show some patients with localized CDC can have very long relapse free survival despite the aggressive nature of this tumor. The two survivors had T1N0M0 Grade 3 and T3bN0M0 grade 4 disease. One patient with T3bN0M0 did not survive but I find it encouraging that even a with a large tumor long term survival is possible.

The other 7 patients had either positive lymph nodes or distant metastasis and most had large tumors and were symptomatic at diagnosis.

Milowsky MI, Rosmarin A, Tickoo SK, Papanicolaou N, Nanus DM.
Active chemotherapy for collecting duct carcinoma of the kidney: a case report and review of the literature.
Cancer. 2002 Jan 1;94(1):111-6. [PubMed Abstract (will open in new window)]

Comment: Although this patient responded dramatically to chemotherapy, the response was very brief indeed. This is at best a marginal suggestion that this chemotherapy could be useful in collecting duct cancer.

This CancerGuide Page By Steve Dunn. © Steve Dunn
Page Created: June 15, 2003, Last Updated: June 12, 2004