In metastatic renal cell carcinoma, pancreatic metastases can appear in two clinical manifestations: (a) very rarely as isolated pancreatic metastases and (b) in the context with multi-organ metastatic disease. Both courses are characterised by rare, unusual clinical features.
Metastatic cancer to the pancreas from other primary sites is rare. In autopsy studies, the frequency is estimated at 1.6–11% [1][2][3][4]. Of 16,614 pancreatic operations due to malignancy listed in 16 clinical reports [5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20], only 258 were performed for pancreatic metastases (PM), resulting in a frequency of just 1.6%. Renal cell carcinoma (RCC), lung carcinomas, colorectal malignancies, melanomas, and sarcomas were reported as the most common primary malignancies [1][2][7][21][22][23][24]. With a proportion of over 60%, RCC is by far the most common primary tumour leading to PM [1][7][14][16][21][22][24][25][26][27], with PM being more frequent in the clear-cell subtype [28]. In the course of metastatic RCC (mRCC), PM may appear: (a) as the sole metastatic site:isolated pancreatic metastasis of RCC (isPMRCC) and (b) in the context with multi-organ metastatic disease.
In older literature there are reports that with synchronous and especially multiple isPMRCC a surgical therapy is not meaningful, since these—analogous to the experience with other solid tumours—were regarded as a negative prognosis criterion with a poor overall survival (OS) [29][30][31][32][33]. In a literature compilation in 2006 [34], our working group demonstrated, however, that the treatment results are not influenced by the singular or multiple or by the synchronous or metachronous occurrence of the PM. The current literature search revealed 12 large (n > 20) institutional reports and compilations [8][13][16][25][35][36][37][38][39][40][41][42] in which data on the impact of clinical risk factors on the OS of operated isPMRCC were presented (Table 2). (The institutional reports were all retrospective and spanned multi-year periods (15–37 years). All studies were performed in surgical departments and inclusion criteria were uniformly surgically treated isPMRCC. Information on the histological subtype was given in four papers: a total of 144 clear cell tumours were compared with only three papillary and one chromophobe. Detailed gene expression analysis and mutation profiles have not yet been presented in isPMRCC).
Table 2. isPMRCC—Impact of potential risk factors for overall survival (OS) after pancreatic surgery; (n = number of patients; Sing/Mult = singular vs. multiple pancreatic metastases (PM), Syn/Metachr = synchronous vs. metachronous; DFI = disease-free interval between RCC and PM surgery; Peripancr. LN = peripancreatic lymph nodes; n.s. = non-significant; p = p-value).
| Authors | Year | n | Sing/Mult | Size | Syn/Metachr | DFI | Peripancr. LN |
|---|---|---|---|---|---|---|---|
| Institutions reports | |||||||
| Milanetto [37] | 2020 | 36 | n.s.; p 0.61 | n.s.; p 0.30 | p 0.01 | n.s.; p 0.96 | p 0.005 |
| Di Franco [36] | 2020 | 21 | n.s.; p 0.391 | n.s.; p 0.569 | n.s.; p 0.143 | n.s.; p 0.07 | |
| Anderson [25] | 2019 | 29 | n.s. | ||||
| Benhaim [8] | 2015 | 20 | n.s. | ||||
| Schwarz [39] | 2014 | 62 | n.s.; p 0.9 | n.s.; p 0.73 | p 0.009 | ||
| Tosoian [40] | 2014 | 42 | n.s.; p 0.727 | n.s.; p 0.602 | n.s.; p 0.509 | n.s.; p 0.738 | n.s.; p 0.085 |
| Konstantinidis [13] | 2010 | 20 | n.s.; p 0.87 | n.s.; p 0.87 | |||
| Reddy [16] | 2008 | 21 | n.s.; p 0.13 | n.s.; p 0.98 | p 0.01 | ||
| Literature compilations | |||||||
| Sellner [35] | 2020 | 527 | n.s.; p 0.350 | n.s.; p 0.423 | n.s.; p 0.790 | n.s.; p 0.786 | |
| Dong [38] | 2016 | 193 | n.s.; p 0.67 | n.s.; p 0.87 | n.s.; p 0.91 | n.s.; p 0.53 | p 0.03 |
| Masetti [42] | 2010 | 157 | n.s.; p 0.862 | n.s.; p 0.092 | p 0.017 | ||
| Tanis [41] | 2009 | 311 | n.s.; p 0.892 | n.s.; p 0.894 | n.s.; p 0.528 | n.s.; p 0.528 | n.s.; p 0.85 |
The influence of single vs. multiple PM on the OS was calculated in 11 studies, all of which yielded a negative result. The influence of tumour size was documented in eight publications, all of which were also negative; the disease-free interval (DFI) between RCC and PM surgery was analysed eight times, with one exception always with a negative result. Finally, the influence of synchronous vs. metachronous PM was investigated in seven papers, six of which yielded a negative result. Of a total of 34 analyses investigating the influence of clinical risk factors such as multiple PM, size, synchronous PM, and DFI, 94.1% thus produced a negative result. The analysis of peripancreatic lymph node involvement revealed a contrary result. Investigated in seven publications, an impact on the OS was identified four times. These LN metastases apparently indicate the beginning of the transition from isPMRCC to multiple organ site metastases with poor prognosis in a relevant proportion of observations. Finally, a possible influence of staging and grading of the primary tumour has been investigated so far only by Di Franco [36], who reported a lack of influence. A study of the influence of the IMDC score [43] or molecular markers in isPMRCC was not found in the cited works.
In summary, this confirms that clinical prognostic factors that reflect tumour burden and tumour growth rate, such as size, occurrence of multiple metastases, DFI, or synchronous occurrence of PM, which are generally relevant in metastasis surgery of solid carcinomas [44][45][46], remain ineffective with isPMRCC. In particular, the complete lack of influence of the tumour volume on survival contradicts a general concept of cancer, in which a greater tumour burden of the body results in a poorer outcome [47].
In early studies, PM in multi-organ metastases of RCC were often classified as terminal events, with poor forecast due to the lack of an effective drug therapy [48]. Only advances in the drug therapy of the RCC—the introduction in clinical praxis of targeted therapies with antiangiogenetic agents and immunotherapy [49][50][51]—have increasingly sparked interest in the actual frequency, the course, and the clinical relevance of these PM and initiated further investigations. The results (Table 3) unexpectedly show that patients with multi-organ site mRCC and concomitant PM have more favourable survival times than those without PM [52][53][54][47][51]. (The design of the five studies was retrospective four times and was unspecified once. The inclusion criteria were three times a first line anti-angiogenetic therapy, once all PM diagnosed in the observation period, and once were not specified. Detailed genetic and molecular markers analyses have only been reported once [53]).
Table 3. Site metastatic RCC with and without PM and survival; (N = size of the PM group; n = number of metastatic organ sites in the PM group and in control (non-PM) group).
| Author, Year | N | Number (n) of Affected Organs and Frequency (%) |
Median Survival (Months) Control vs. PM Group |
||
|---|---|---|---|---|---|
| n | Control Group | PM Group | Significance | ||
| Singla [53], 2020 | 31 | 35 vs. 101 | |||
| p < 0.001 | |||||
| Chrom [47], 2018 | 34 | 1–2 | 52% | 38% | 23 vs. 46 |
| 3, >3 | 48% | 62% | p = 0.022 | ||
| Kalra [51], 2016 | 44 | 1 | 31% | 9% | 26 vs. 39 |
| 2 | 40% | 23% | p < 0.01 | ||
| 3, >3 | 29% | 68% | |||
| Yuasa [54], 2015 | 20 | p < 0.0001 | |||
| Grassi [52], 2013 | 24 | 1 | 37% | 25% | 23 vs. 39 |
| 2 | 35% | 21% | p = 0.0004 | ||
| 3, >3 | 28% | 54% | |||
In 2013, Grassi [52] reported for the first time that in mRCC the median OS was more favourable for patients with PM at 39 months than for those without PM at 23 months (p = 0.0004), an observation made by Yuasa [54] in 2015 reaffirmed. Kalra [51] confirmed this result in a comprehensive study again in 2016: a median OS of 39 months for patients with PM corresponded to 26 months for patients without PM (p < 0.01), although the total number of affected organs was greater in the collective with PM (p < 0.001). Chrom [47] also observed a better OS for the group of mRCC with pancreatic involvement: 46 months vs. 23 months, which was significant in univariate analysis (p = 0.022) but did not prove to be an independent variable in multivariate analysis. In a recent study, Singla [53] compared mRCC with and without presence of PM with regard to clinical behaviour and underlying biology. In doing so it was reaffirmed that the cohort with PM had a more favourable OS than the one without PM (OS 101 vs. 35 months (p < 0.001), 5-year survival rates 88% vs. 31%, p < 0.001). There was also a high sensitivity of the cohort with PM to antiangiogenetic agents with a disease-free survival of 26.9 vs. 8.3 months (p = 0.007) and a simultaneous resistance to immune checkpoint inhibitors with more rapid progress of PM observations (p = 0.034). Finally, within the PM cohort, Singla noted an independence (p = 0.684) of survival from the established IMDC score, which is not based on pathological–anatomical parameters. This observation, however, is in contrast to Yuasa [54], who observed in the PM group a significant difference in OS between good and intermediate IMDC risk patients (p = 0. 013). As a summary, all five investigations carried out so far consistently show that the presence of PM exerts a positive influence on the OS in mRCC.
However, in the three publications [52][47][51] (Table 3) that reported on the possible influence of the number of organs affected by metastases, it was consistently shown that the proportion of single organ metastases was smaller in mRCC with PM than in the respective non-PM control group, and vice versa, the rate of observations with three or more affected organs in the PM group was higher than in the control group. The cohorts with PM were thus consistently characterised by a reduced proportion of prognostically favourable single organ metastases and an increased proportion of prognostically unfavourable multi-organ metastases. In the study of Kalra [51], the difference even reached significance level (p < 0.001). Thus, the paradox is that in the mRCC, the presence of PM is associated with more favourable OS, although the total number of affected metastatic organ sites is greater. Therefore, an unconscious selection bias of favourable cases with lower number of metastatic organ sites in the PM group has to be excluded as the cause of the more favourable prognosis. The better prognosis, despite the higher rate of affected organs in the PM group, indicates that the occurrence of PM in the mRCC is associated with tumour cell properties, which trigger a more favourable course [52][53][51].
The occurrence of PM in metastatic RCC—both isPMRCC and PM in multiple organ site mRCC—is coupled to clinical features that differ in some aspects from those common in solid tumour metastases, e.g., the possibility of isolated occurrence (isPMRCC), ineffectiveness of volume and growth-rate related risk factors, and better prognosis despite a larger number of affected organs. In the case of isPMRCC in particular, three special features can be detected that differ from the behaviour generally observed in extrapancreatic metastases of mRCC: (1) Genetic alterations associated with lower tumour aggressiveness; (2) The lack of influence of tumour volume and growth rate dependent prognostic factors; and (3) Immunotherapy, which is very effective in mRCC, remains ineffective in isPMRCC. Of course, it is conceivable that these three properties are completely independent of each other. However, this leads to the unsatisfactory explanation that these embolised RCC cells are equipped with an accumulation of unusual biological properties. The data presented, however, can just as well be explained by the coincidence of special RCC cells with a postulated highly specific SSM in isPMRCC. At the beginning of the process there is undoubtedly the evolution of tumour cells that leads to a multiplicity of different cell species with different cell genomes (according to the cancer evolutionary model, this cell diversity is the result of a phase of genome chaos [55][56][57][58] with following genome re-organisation [59][60][58], which is responsible not only for a diversity of cell genomes, but also for possible chromosomal rearrangements that are associated with drug resistance [56][61], drawing attention to the concept of adaptative therapy [59][56][62]). It is a specificity of the isPMRCC entity that those few circulating tumour cells that eventually settle and develop into manifest metastases are linked with reduced aggressiveness, as evidenced by genetic studies as well as the clinically proven slow progression of the disease. However, the reduced aggressiveness by itself cannot explain the phenomenon of isolated occurrence of PM and the lack of influence of risk factors. According to current knowledge, an organotropism in metastasis development—and one such is undoubtedly the isolated occurrence of PM in mRCC—is linked to a SSM. Therefore, the exclusive occurrence of metastases in one organ (pancreas) presupposes that these cells, during their evolution, are endowed not only with a low aggressiveness but also with additional properties that can trigger a highly selective SSM in host organs. This favours metastasis settlement in individual organs and prevents it in others. Therefore, in isPMRCC too, the clinical behaviour is the common end product of tumour and host biology [63].
Which biochemical alterations of the tumour cells and/or of the host organs trigger such a SSM in isPMRCC is currently unknown and reserved for future investigations. Though isolated PM of the RCC are very rare, they deserve increased investigations as there are indications that clinical course is determined—in addition to genetic changes— to an unusually high degree by a SSM too. An elucidation of the biochemical changes that control this SSM could lead to a deeper understanding of the processes that promote or prevent the colonisation of metastatic tumour cells.
This entry is adapted from the peer-reviewed paper 10.3390/cancers13061342