Colorectal Liver Metastases: Does the Future of Precision Medicine Lie in Genetic Testing?
- 217 Downloads
Colorectal liver metastases (CRLM) present an important clinical challenge in both surgical and medical oncology. Despite improvements in management, survival among patients undergoing resection of CRLM is still very variable and there is a paucity of clinical trial data and reliable biomarkers that could guide prognostic forecasts, treatment selection, and follow-up. Fortunately, recent advances in molecular biology and tumor sequencing have identified a number of critical genetic loci and proliferation markers that may hold the key to understanding the biologic behavior of CRLM; specifically, mutations of KRAS, BRAF, TP53, PIK3CA, APC, expression of Ki-67, and the presence of microsatellite instability appear to have a decisive impact on prognosis and response to treatment in patients with CRLM. While the applicability of genetic biomarkers in everyday clinical practice remains conditional on the development of inexpensive bedside sequencing, targeted therapies, and the conduct of appropriate clinical trials, the promise of personalized treatment may be closer to realization than ever before.
KeywordsColorectal liver metastases Genetics Precision medicine
Colorectal cancer (CRC) is the third most common malignancy in the world and the third leading cause of cancer-related death in the USA.1 CRC commonly invades the liver through the portal circulation; in fact, the liver is the most common site of metastatic disease and the presence of liver metastasis is a major determinant of prognosis in this patient population.2, 3, 4, 5, 6 Hepatic resection is the accepted standard of care for colorectal liver metastasis (CRLM), with a 5-year survival rate that ranges from 25 to 58%, compared to 10–15% for patients with unresectable disease.7, 8, 9, 10, 11 However, as many as 60–70% of surgically resected patients eventually recur.12,13
Given this high propensity for recurrence, treatment selection can be challenging for patients with colorectal liver metastasis.14 To this point, the possible role of biomarkers as guides to both prognosis and treatment is presently being explored. Historically, clinicopathologic features like primary tumor stage, carcinoembryonic antigen (CEA) level, nodal status, number and size of liver metastases, resection margin status, presence of extrahepatic disease, and interval between primary tumor diagnosis and liver metastasis have been used to determine prognosis among surgical candidates.12 However, the prognostic power of these variables has been questioned, with a number of recent studies demonstrating only a poor and inconsistent correlation between traditional clinicopathologic factors and outcomes.15 To address these limitations, contemporary research is focused on exploring the possible role of genetic factors in the prognosis of patients with CRLM.14, 15, 16 In particular, the role of mutations of the Kirsten rat sarcoma viral oncogene homolog (KRAS) (30–45% of patients with CRC), as well as of phosphoinositide-3 kinase (PIK3CA; 16%), V-raf murine sarcoma b-viral oncogene homolog B1 (BRAF; found in 3 to 11% of CRC patients) and NRAS (evidenced in 2.6 to 6.2% of patients)17,18 has attracted significant attention in the literature while the role of other somatic mutations (e.g., MSI, APC, and TP53) has not been well studied. In the present review, we will attempt to summarize the most important relevant findings and discuss their implications for contemporary clinical practice.17, 18, 19, 20, 21, 22
A number of critical genetic loci and proliferation markers
Prognostic relevance in CRLM
Potential targeted therapies
GTP-binding protein that once activated transfers signals from the cell membrane to the nucleus, through mitogen activated protein kinases (MAPK).
Generally accepted as a marker of aggressive disease.
Resistant to therapies with anti-EGFR.
Kemeny et al.13
Brudvik et al.20
Cejas et al.24
Samatar et al.25
Vauthey et al.32
Nash et al.33
Petrowsky et al. 34
Rose et al.35
Andreatos et al.38
Andreyev et al.41
Imamura et al.42
Spolverato et al.43
Part of the MAPK cascade, downstream from KRAS.
Multiple studies confirm its negative prognostic impact.
Possible resistance to anti-EGFR therapy.
ERK inhibitors could be important components against mutBRAF in CRC. Early clinical trials are being conducted.44
Tran et al.22
Teng et al.45
Tie et al.46
Souglakos et al.47
Karagkounis et al.50
Bokemeyer et al.51
Schirripa et al.52
Margonis et al.53
TP53 is a tumor suppressor gene and therefore it is involved in the regulation of the cell cycle.
Conflicting results over its prognostic impact; possibly its impact may be influenced by other mutations and therapies.
Clinical trials are underway with compounds that induce reactivation of p53 or that deplete mutant p53.54
Chun et al.55
Pilat et al.58
It is involved in cell proliferation and survival by encoding the catalytic subunit of PI3K kinase.
Only two studies assessed its impact on CRLM; they found a decreased time to relapse after surgery in mutPIK3CA and a correlation with negative survival when concomitantly present with APC.
Possible resistance to anti-EGFR therapy.
PIK3CA inhibitors are currently being evaluated in clinical trials as novel anti-cancer therapies.61
De Rook et al.17
Souglakos et al.47
Yamashita et al.62
Samuels et al.63
Spolverato et al.43
Li et al.64
Sartore-Bianchi et al.65
Prenen et al.66
Loes et al.67
Nuclear protein linked to cellular proliferation
Conflicting results. Some studies link its increased expression to high proliferation index and negative prognosis after resection, while others found an improved survival.
Studies are underway to selectively target Ki-67, by killing cells expressing it. However research in this field is only at its premature stages.68
Nash et al.33
Petrowsy et al.34
Ivanecz et al.56
Tumor suppressor gene and therefore it is involved in the regulation of the cell cycle. Its loss leads to the activation of WNT signaling pathway.
Conflicting results on whether its concomitant presence with PIK3CA is prognostic.
Efforts have been made to develop agents targeting WNT signaling pathway, however clinically relevant inhibitors are not available yet.69
Chun et al.55
Yamashita et al.62
Repeating units present in the DNA.
Do not seem to be prognostic; however, its presence is high in patients with BRAF mutations making its individual role difficult to estimate.
Pembrolizumab, an anti-programmed death 1 immune checkpoint inhibitor seems to provide a clinical benefit for patients with dMMR.44
Tran et al.22
Vilar et al.70
Popat et al.71
Goldstein et al.72
Koopman et al.73
Muller et al.74
Kochhar et al.75
Haddad et al.76
Chen et al.77
A recent meta-analysis reported a pooled KRAS mutation rate of 30.6% among patients with CRLM.20 The role of KRAS mutation among patients with CRLM has been extensively studied.13,20,23,32 It is generally accepted that KRAS mutation is a marker of aggressive disease, resulting in earlier and more frequent recurrence after metastasectomy.20,32 In turn, studies from mouse models suggest that the impact of KRAS mutations on outcomes may be partly mediated by effects on the actin cytoskeleton, which promote metastatic spread and are crucial for the maintenance of a transformed phenotype.79 Consequently, KRAS mutation status has been proposed as a predictive marker for prognosis after liver resection in CRLM.33 Nonetheless, some studies have failed to detect a correlation between KRAS mutation and prognosis.34,35,45 As such, a brief review of the published evidence may prove illuminating.
Kemeny et al. found that in patients who underwent liver resection for metastases, KRAS was a significant predictor of 3-year recurrence-free survival (RFS) (HR = 1.9; P = 0.01), with higher recurrence rates to the lung, brain, and bone compared to wild-type KRAS (wtKRAS).13 In another prospective study of 633 patients undergoing resection for CRLM, RAS was associated with both a worse 5-year overall survival (HR = 1.629; P = 0.044) and a higher incidence of positive margins.36 Similarly, Vauthey et al. in a cohort of patients undergoing hepatectomy for CRLM found that patients with mutRAS genotype (KRAS/NRAS) had a 3-year overall survival of 52.2 versus 81% for wtRAS (P = 0.002); 3-year RFS was 33.5% in patients with a wtRAS genotype versus 13.5% for mutRAS (P = 0.001). Interestingly, the presence of RAS mutation increased the risk of recurrence to the lung (HR = 2.0; P = 0.01), but not to the liver.32 Moreover, a comprehensive meta-analysis (14 studies; 1800 patients) involving patients who underwent resection for CRLM demonstrated a negative association of mutKRAS status with both overall survival (OS) (HR = 2.24) and RFS (HR = 1.89).20 Although four of the studies included in the meta-analysis failed to report a relationship between KRAS mutational status and overall survival, the pooled results clearly suggest that KRAS is a powerful predictor of recurrence and death among surgically treated patients with CRLM.13,24,34,38,45 Interestingly, the presence of KRAS mutation among patients who underwent recurrence after liver resection may herald an even worse prognosis, according to a recent study by Margonis et al.23 In turn, these findings suggest that far from merely increasing the rate of recurrence, mutKRAS genotype may be associated with more aggressive recurrence patterns that may be more difficult to treat. Future studies are needed to further assess this hypothesis.
In spite of the strong evidence in favor of a uniform prognostic role for KRAS mutational status, some studies suggest that patient-specific factors may greatly modify outcomes. For example, in patients undergoing liver surgery for CRLM, Sasaki et al. recently demonstrated that while mutKRAS status was associated with worse 5-year RFS and OS (wtKRAS = 23.7 and 57.2% vs. mutKRAS = 19.6 and 38.2%, respectively; both with P = 0.05), among patients with left-sided primary tumors (HR = 1.57; P = 0.04 and HR = 1.81; P = 0.02, respectively), this was not the case among patients with right-sided tumors.39 Given the established differences in pathophysiology and clinical characteristics between left- and right-sided CRC in both the primary and metastatic settings, these findings suggest that the impact of KRAS mutations may differ sharply between colorectal tumors with different genetic backgrounds.39,40
Aside from patient-specific factors, the specific codon and point mutations affecting the KRAS gene locus are also thought to have different prognostic implications.27 A study conducted on over 3400 patients with colorectal cancer and KRAS mutation showed unequivocally that different gene mutations, even in the same site of the genome, have different influences on outcome and that certain mutations are associated with a more aggressive behavior in advanced colorectal cancer.41,42 For example, a report on 1075 patients with CRC demonstrated that patients with KRAS codon 12 mutations fared worse than those with codon 13 mutations.42 Similarly, a separate study on patients undergoing CRLM resection demonstrated that codon 13 mutations were not associated with worse 5-year OS (P = 0.13). Furthermore, they identified two mutations in codon 12, G12V (HR = 1.78; 95% CI, 1.00–3.17; P = 0.05) and G12S (HR = 3.33; 95% CI, 1.22–9.10; P = 0.02) that accounted for up to 70% of the increased risk of death compared to wtKRAS (HR = 1.70; 95% CI, 1.13–2.55; P = 0.01). These effects were even more pronounced in the subpopulation of patients who recurred after hepatic resection, with G12V (HR = 2.96; 95% CI, 1.32–6.61; P = 0.01), G12C (HR = 6.74; 95% CI, 2.05–22.2; P = 0.002), and G12S mutations (HR, 4.91; 95% CI, 1.52–15.8; P = 0.01) being associated with worse 5-year OS (P < 0.05).27 Nonetheless, codon 13 mutations are not entirely benign. According to later studies on patients that underwent surgery for CRLM, they do appear to negatively affect extrahepatic 5-year RFS (HR = 2.27, 95% CI, 1.29–3.97; P = 0.004) and 5-year lung-specific RFS (HR = 2.32, 95% CI, 1.12–4.78; P = 0.023), thus increasing the risk of extrahepatic recurrence by 69.4% and of lung recurrences by 70% compared to patients with codon 12 mutations.28
The management of patients with mutKRAS status and CRLM remains controversial. No systemic treatments that target KRAS are currently available, but relevant research is underway.25 With respect to surgical therapy, it has been suggested that, as patients with mutKRAS tumors are more likely to have positive resection margins on pathology, wider margins of resection (1.5 cm for resection with ultrasound guidance) and/or ablation (1 cm with CT guidance) may lead to better outcomes.36,37 Although this suggestion is reasonable, there is currently little direct evidence that wider resection margins lead to improved outcomes in this patient group. On the contrary, a recent retrospective study on CRLM resection by the Johns Hopkins Hospital (JHH) group suggests that margin width may have, at best, only a limited impact on the prognosis of patients with mutKRAS status; in fact, while an R0 resection was associated with improved survival among patients with wtKRAS lesions, no significant survival advantage was noted among patients with mutKRAS tumors.29,30 In turn, these findings suggest that the increased incidence of positive margins among mutKRAS patients may not be the root cause of adverse long-term outcomes, but merely a “symptom” of the infiltrative local and systemic growth pattern of mutKRAS lesions; this finding has already been described in detail in patients with lung adenocarcinoma.80 In turn, wider resection alone may be an ineffective therapeutic strategy, if it is not linked to more comprehensive efforts to limit disease spread. While little can be done to prevent systemic spread on the basis of surgery alone, local/intrahepatic spread may perhaps be controlled by preemptively eliminating the route of metastatic spread through the portal tributaries and lymphatics, with the aid of an anatomical resection (AR). Of note, while anatomical resections are considered the standard of care in hepatocellular carcinoma, they have not previously been shown to confer a prognostic benefit in CRLM.81,82 However, the first study to assess the utility of anatomical resections among patients with mutKRAS CRLM concluded that anatomical resections confer a striking improvement in overall RFS: 5-year RFS was 14.4% in the non-anatomical resection (non-AR) arm, compared to 46.4% in the AR arm (HR = 0.45; P = 0.002); similar results were obtained for intrahepatic RFS (78.8 in the AR vs 10.3 months in the non-AR, P < 0.001; HR = 0.42. 95% CI, 0.23–0.78; P = 0.06). On the contrary, wtKRAS patients obtained no benefit from anatomical resection, in line with what has been previously reported for unselected cohorts (including both mutKRAS and wtKRAS patients).31
BRAF is part of the mitogen-activated protein kinase cascade (MAPK), downstream from KRAS. Interestingly, sporadic colorectal cancers with microsatellite instability (MSI) have a high incidence of mutated BRAF (mutBRAF), ranging from 40 to 60%, whereas the same mutation is found in only 5–10% of microsatellite stable tumors.46 The most common BRAF mutation encountered in colorectal cancer is V600E.83 BRAF mutations are a hallmark of aggressive disease for patients with metastatic colorectal cancer, with decreased survival and significantly lower RFS with respect to the wild-type counterpart, while resistance to anti-epidermal growth factor receptor (EGFR) therapy is also a possible concern.47 In a pooled analysis of more than 22,600 patients with metastatic colorectal cancer, BRAF was identified as a contributing factor to 90-day mortality.48
A recent meta-analysis reported a pooled BRAF mutation rate of 2–4% in patients with CRLM.49 This very low incidence of mutBRAF in patients that undergo CRLM resection may be explained by the high frequency of aggressive, unresectable disease associated with this mutation.50 Multiple studies confirm the adverse prognostic impact of BRAF mutations in patients with colorectal liver metastases with worse median OS compared to wtBRAF (10.4 vs. 34.7 months, P < 0.01); some suggest that BRAF might actually be a more powerful predictor of prognosis than RAS in patients undergoing resection for CRLM.22,45,46,50,51 Teng et al. found that in patients with liver-limited metastases who have undergone liver metastasectomy, BRAF but not KRAS was an independent prognostic biomarker for worse 5-year OS (HR = 6.243, 95% CI, 1.89–20.591; P = 0.003).45 Similarly, Schirripa et al. in their multivariable analysis identified BRAF patients undergoing liver resection for CRLM, as being at higher risk of relapse with respect to all wild-type (HR = 2.31, 95% CI, 1.09–4.87; P = 0.029) and mutKRAS patients (HR = 2.06, 95% CI, 1.02–4.14, P = 0.044).52
Nonetheless, a recent study by the JHH group suggests that the impact of BRAF mutations on survival in patients with surgically treated CRLM is most apparent during the first postoperative year (HR = 2.22, 95 CI%, 1.40–3.51; P = 0.001); afterwards, BRAF mutational status ceases to be a prognostic factor (3 years post-surgery, P = 0.260).53 Comparing the specific BRAF mutations carried by the survivors with those of patients that died during the first postoperative year might shed further light on these findings.
TP53 is a tumor suppressor gene that plays a crucial role in the regulation of the cell cycle; it induces apoptosis and aids in DNA repair. As such, its close association with cancer development, control, and progression is unsurprising. The incidence of TP53 mutation in patients with CRLM has been reported to range between 48 and 65.6%.55, 56, 57, 58, 59, 60 As early as the 1990s, a study that investigated the prognostic impact of TP53 mutation in patients with CRLM after hepatic resection identified it as a predictor of poor survival (P = 0.0079), even after resection of hepatic metastases.57,59 In clear contrast to these studies, however, Yang et al. reported improved survival after surgery among patients with CRLM and TP53 mutations.60 On the other hand, Ivanecz et al. reported no association between the presence of TP53 mutation and survival in the resected CRLM population.56
Although these conflicting results have yet to be satisfactorily explained, it is possible that the impact of TP53 mutation on outcomes may vary depending on interaction with other genetic mutations and/or treatment setting. For example, a recent study by Chun et al. reported strikingly worse survival among patients undergoing hepatic surgery for CRLM with concomitant RAS and TP53 mutations (identified in 31.4% of the patients); on the contrary, isolated RAS and TP53 mutations were not shown to be prognostic by themselves in multivariable analysis.55 Interestingly, another study targeting patients undergoing CRLM resection demonstrated that TP53 mutation was only associated with a worse 5-year overall survival among patients who received neoadjuvant chemotherapy; no such association was noted among patients treated with surgery alone.58 Given the propensity of chemotherapy to induce de novo mutations and what is known about the function of TP53, it is possible that TP53 mutations essentially serve a permissive role, accelerating the acquisition of additional “driver” mutations that confer a more aggressive phenotype and worse prognosis. In turn, this might explain why the presence of TP53 mutations does not, by itself, lead to worse outcomes. Additional research is needed to elucidate the pathogenic role of TP53 in patients with CRLM.
This gene is mutated in approximately 15% of CRC cases.17,18,47,55 It encodes the catalytic subunit of a kinase called PI3K (phosphoinositide-3 kinase), which is involved in cell proliferation and survival.62 PIK3CA mutations result in attenuation of apoptosis, facilitate tumor migration and invasion, and commonly occur during the transition from large adenoma to cancer.43,63 Interestingly, the presence of these mutations seems to favor lung (20%) and brain (23.9%) over liver metastases (7.7%).18 To the contrary, CRC patients with KRAS and PIK3CA double-mutation are more likely to develop liver metastases.64
Compared with wild-type PIK3CA (wtPIK3CA), the presence of PIK3CA mutations in the metastatic CRC population (mCRC) does not seem to directly impact 3-year RFS or OS.17 However, in the absence of KRAS mutations, mutated PIK3CA (mutPIK3CA) status has been linked with resistance to anti-EGFR agents in patients with mCRC, possibly accounting for up to 17% of such cases.65 Similarly, Souglakos et al. reported an increased risk of relapse and shorter RFS after anti-EGFR therapy in the presence of PIK3CA mutations; nonetheless, in the mCRC population, the reported effects were limited in comparison to KRAS and BRAF mutations.47 However, other studies on the same type of population (mCRC) have failed to replicate these findings.66 While some authors attribute these disparate results to differences in biologic behavior between various PIK3CA mutations (exon 20 mutations are thought to underlie resistance to anti-EGFR agents, while exon 9 mutations appear more biologically indolent), the frequent concomitant occurrence of PIK3CA, KRAS, and BRAF mutations renders interpretation challenging.17
The incidence of PIK3CA mutation in patients with CRLM has been reported to range between 13.4 and 20.9%.62,67 To the best of our knowledge, there are only two studies assessing the impact of PIK3CA in patients undergoing hepatic resection for CRLM: Løes et al. associated the presence of mutPIK3CA to a decreased time to relapse with respect to the wild-type PIK3CA (4 vs. 17 months; P = 0.023), but failed to find the same association with regard to the disease-specific survival (P = 0.174).67 The other study, by Yamashita et al., found a correlation between the concomitant presence of PIK3CA and adenomatous polyposis coli (APC) with 3-year OS and RFS and is reported under the APC paragraph.62
Ki-67 is found in all active phases of the cell cycle and drives cellular proliferation. This protein’s prognostic significance for survival and recurrence has been previously demonstrated in breast and prostate cancer.84 High Ki-67 expression has been reported in 19.5 to 62% of patients with CRLM.33,34,56 In CRC and CRLM, its increased expression is associated with a high proliferation index, which in turn serves as a negative prognostic factor: patients with high scores (≥ 50%) have a lower median survival after curative hepatic resection.34 In fact, a recent study from the Memorial Sloan Kettering Cancer Center on patients with resected CRLM demonstrated that high Ki-67 expression may be an even stronger predictor of prognosis than KRAS; interestingly, patients with high Ki-67 expression were more likely to present with synchronous CRLM, high tumor burden, and preoperative CEA > 200 ng/ml and less likely to undergo curative intent resection. Interestingly, the impact of high Ki-67 expression was evident in the group undergoing colon resection only, but appeared limited among those who successfully underwent complete resection of both primary CRC and CRLM.33 Precisely, the opposite trend was demonstrated in a retrospective analysis by Ivanecz et al., which linked Ki-67 overexpression with improved survival in those who underwent resection for CRLM.56 The clinical significance of these conflicting reports remains under investigation.
One of the initial steps in colon cancer formation is the development of polyps, whose growth is triggered by inherited or somatic loss-of-function mutations in the tumor suppressor gene adenomatous polyposis coli (APC).85 APC mutations are present in 50–83% of sporadic colorectal cancers.86
The incidence of APC mutation in patients with CRLM has been reported to range between 47.4 and 48.7%.55,62 A recent study from the MD Anderson Cancer Center assessed the prognostic impact of concomitant APC and PIK3CA mutations in patients undergoing hepatectomy for colorectal liver metastases, as the two frequently co-occur in patients with colorectal cancer. The authors reported that while none of the two mutations was individually prognostic, their concomitant presence had a severe impact on 3-year OS (44 vs. 84%; P < 0.001) and 3-year RFS (3.1 vs 20%; P < 0.001) with respect to the whole cohort.62 However, a similar study on CRLM-resected patients failed to confirm the independent impact of concomitant APC and PIK3CA mutations on survival; further studies are needed to shed light on this issue.55
Microsatellite sequences are repeating units present in the DNA; when insertions and deletions occur in these DNA sections, they are repaired by the mismatch repair system (MMR), composed of several proteins like MLH1, MSH2, MSH6. When this system is defective, it leads to deficient MMR (dMMR), which promotes mutagenesis and predisposes to tumor formation. Patients can be grouped into three distinct categories with respect to MSI status: high microsatellite instability (MSI-H), low microsatellite instability (MSI-L), and microsatellite stable.70 The presence of 30% of microsatellite markers for MSI-H or the absence of one of the MMR proteins determines classification in the MSI-H category.70 MSI-H tumors have distinct features such as predilection for the proximal colon, poorer differentiation, mucinous histology, and pronounced lymphocytic infiltration.71
These mutations can either be encountered in the context of the autosomal dominant hereditary non-polyposis colorectal cancer (HNPCC) syndrome (Lynch syndrome) or they can be sporadic due to MLH1 promoter hypermethylation. Sporadic mutations are more commonly found in CRC (10–20%).72,73 Deficient MMR is rare in advanced sporadic tumors, making it difficult to draw conclusions on the connection between MMR status and advanced stage cancers. Microsatellite instability is more common in stage II (20%) rather than III (12%) and is even rarer in stage IV (4%), suggesting that these tumors have a reduced metastatic potential.70,73 Probably owing to the low incidence of MSI-H in mCRC, very few studies have assessed its role in metastatic disease. Specifically, Müller et al. investigated 108 patients with metastatic disease who were part of a prospective randomized trial comparing oxaliplatin and fluorouracil (FU) and found a low incidence of MSI-H (4%). Interestingly, the authors were able to identify a lower rate of disease control for MSI-H, who responded to FU/oxaliplatin in only 50% of the cases, compared to 95.6% for non-MSI-H patients (P = 0.002). Nonetheless, this difference in response did not translate to a significant difference in 3-year RFS or OS.74 An extensive review including 32 studies and over 7600 cases aimed to investigate the association between MSI and prognosis in CRC patients, identified MSI in 1277 patients. Interestingly, this study concluded that the presence of MSI conferred a significantly better prognosis with respect to patients whose mismatch repair mechanisms were intact.71 The incidence of MSI among patients with CRLM has been reported to range between 2.5 and 43.3%. However, these results are based on very few studies, none of which has reported an association of MSI status with survival and all of them failed to find an association with survival.75, 76, 77 Specifically, Kochlar et al. and Haddad et al., in two different studies, identified MSI in only 2.5% of the cohort and did not find it to be prognostic.75,76
As previously described, in CRC, the frequency of BRAF mutation is particularly high in patients with MSI; over 50% of patients with mutBRAF also have MMR-deficient cancers. On the contrary, KRAS mutations occur less commonly in this patient population.87 Unfortunately, the true prognostic impact of MSI is difficult to estimate as BRAF mutations tend to drive the prognosis whenever present in mCRC.22
The outcome of metastatic CRC and particularly CRLM is very variable and, not infrequently, traditional clinicopathologic markers of prognosis fail to predict survival or direct management. The study of tumor genetics promises to radically alter this landscape thus enabling the implementation of truly personalized therapy for the first time.43
The development of targeted therapies, already extremely effective in other cancers, is the most immediate benefit of tumor sequencing. Although targeted agents as effective as, e.g., Imatinib have yet to emerge in CRC, awareness of the important role of KRAS and BRAF mutations has led to the development of inhibitors which have shown promising results in early trials.44 Moreover, determination of the genetic profile of the tumor does not only drive development of novel therapies, but allows for the optimized use of existing ones. In the past, trials of surgical, interventional, and systemic therapy could only be directed on the basis of broad clinical and pathologic criteria and thus included extremely heterogeneous patient populations; as such, therapies that were extremely effective in specific subgroups, but not in the overall cohort, would have likely failed to be identified. Stratification according to genetic profile promises to largely resolve this issue and is an integral part of modern clinical trial design; the identification of MSI tumors as extremely favorable targets for immunotherapy is only the most recent relevant example.88 Similarly, knowledge of the natural history and specific patterns of recurrence associated with given mutations may promote the development of personalized and, consequently, more cost-effective follow-up strategies.13,22 Furthermore, the sequence of treatment (e.g., metastasectomy followed by chemotherapy vs. neoadjuvant chemotherapy followed by surgery) can also be better directed if the natural progression of disease can be anticipated on the basis of genetic characteristics. For example, the presence of the V600E BRAF mutation in patients with CRLM has been associated with rapid disease progression and death within the first postoperative year; in this context, the wisdom of delaying systemic treatment in favor of a localized therapy such as liver resection appears questionable and should perhaps be evaluated by a clinical trial.53 Lastly, even the choice of the appropriate surgical technique (either as anatomical vs non-anatomical hepatectomies or an optimal margin width) might be shaped by the mutational status of the metastatic liver disease.29,31
Personalized therapy on the basis of tumor molecular profiling appears to represent the future of cancer care. While full-scale implementation of these concepts will require development of inexpensive sequencing technologies (the development of “liquid biopsies” on the basis of cell-free DNA is an encouraging step in this direction), refinement of our understanding of tumor genomics and high-quality real-world data from multiple clinical trials, physicians and patients alike have valid reasons for optimism.
Carlotta Barbon: conception and design, drafting and critically revising the article, final approval of the version to be published, agreement to be accountable for all aspects of the work.
Georgios Antonios Margonis: design, critically revising the article, final approval of the version to be published, agreement to be accountable for all aspects of the work.
Nikolaos Andreatos: design, critically revising the article, final approval of the version to be published, agreement to be accountable for all aspects of the work.
Neda Rezaee: design, critically revising the article, final approval of the version to be published, agreement to be accountable for all aspects of the work.
Kazunari Sasaki: design, critically revising the article, final approval of the version to be published, agreement to be accountable for all aspects of the work.
Stefan Buettner: design, critically revising the article, final approval of the version to be published, agreement to be accountable for all aspects of the work.
Christos Damaskos: design, critically revising the article, final approval of the version to be published, agreement to be accountable for all aspects of the work.
Timothy M Pawlik: design, critically revising the article, final approval of the version to be published, agreement to be accountable for all aspects of the work.
Jin He: design, critically revising the article, final approval of the version to be published, agreement to be accountable for all aspects of the work.
Christopher L. Wolfgang: conception and design, critically revising the article, final approval of the version to be published, agreement to be accountable for all aspects of the work.
Matthew J. Weiss: conception and design, critically revising the article, final approval of the version to be published, agreement to be accountable for all aspects of the work.
- 1.Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics, 2009. CA: a cancer journal for clinicians. 2009;59:225–249.Google Scholar
- 7.Abdalla EK, Vauthey JN, Ellis LM, Ellis V, Pollock R, Broglio KR, Hess K, Curley SA. Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Annals of surgery. 2004;239:818–825; discussion 825-817.CrossRefPubMedPubMedCentralGoogle Scholar
- 8.House MG, Ito H, Gonen M, Fong Y, Allen PJ, DeMatteo RP, Brennan MF, Blumgart LH, Jarnagin WR, D'Angelica MI. Survival after hepatic resection for metastatic colorectal cancer: trends in outcomes for 1,600 patients during two decades at a single institution. Journal of the American College of Surgeons. 2010;210:744–752, 752-745.Google Scholar
- 11.American Cancer Society. Colorectal Cancer Facts and Figures 2014–2016, Atlanta: American Cancer Society [Internet] 2014. Available from: https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/colorectal-cancer-facts-and-figures/colorectal-cancer-facts-and-figures-2014-2016.pdf.
- 13.Kemeny NE, Chou JF, Capanu M, Gewirtz AN, Cercek A, Kingham TP, Jarnagin WR, Fong YC, DeMatteo RP, Allen PJ, Shia J, Ang C, Vakiani E, D'Angelica MI. KRAS mutation influences recurrence patterns in patients undergoing hepatic resection of colorectal metastases. Cancer. 2014;120:3965–3971.CrossRefPubMedPubMedCentralGoogle Scholar
- 14.Douillard JY, Oliner KS, Siena S, Tabernero J, Burkes R, Barugel M, Humblet Y, Bodoky G, Cunningham D, Jassem J, Rivera F, Kocakova I, Ruff P, Blasinska-Morawiec M, Smakal M, Canon JL, Rother M, Williams R, Rong A, Wiezorek J, Sidhu R, Patterson SD. Panitumumab-FOLFOX4 treatment and RAS mutations in colorectal cancer. The New England journal of medicine. 2013;369:1023–1034.Google Scholar
- 17.De Roock W, Claes B, Bernasconi D, De Schutter J, Biesmans B, Fountzilas G, Kalogeras KT, Kotoula V, Papamichael D, Laurent-Puig P, Penault-Llorca F, Rougier P, Vincenzi B, Santini D, Tonini G, Cappuzzo F, Frattini M, Molinari F, Saletti P, De Dosso S, Martini M, Bardelli A, Siena S, Sartore-Bianchi A, Tabernero J, Macarulla T, Di Fiore F, Gangloff AO, Ciardiello F, Pfeiffer P, Qvortrup C, Hansen TP, Van Cutsem E, Piessevaux H, Lambrechts D, Delorenzi M, Tejpar S. Effects of KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus chemotherapy in chemotherapy-refractory metastatic colorectal cancer: a retrospective consortium analysis. The Lancet Oncology. 2010;11:753–762.CrossRefPubMedGoogle Scholar
- 18.Tie J, Lipton L, Desai J, Gibbs P, Jorissen RN, Christie M, Drummond KJ, Thomson BN, Usatoff V, Evans PM, Pick AW, Knight S, Carne PW, Berry R, Polglase A, McMurrick P, Zhao Q, Busam D, Strausberg RL, Domingo E, Tomlinson IP, Midgley R, Kerr D, Sieber OM. KRAS mutation is associated with lung metastasis in patients with curatively resected colorectal cancer. Clinical cancer research : an official journal of the American Association for Cancer Research. 2011;17:1122–1130.CrossRefGoogle Scholar
- 21.Knijn N, Mekenkamp LJ, Klomp M, Vink-Borger ME, Tol J, Teerenstra S, Meijer JW, Tebar M, Riemersma S, van Krieken JH, Punt CJ, Nagtegaal ID. KRAS mutation analysis: a comparison between primary tumours and matched liver metastases in 305 colorectal cancer patients. British journal of cancer. 2011;104:1020–1026.CrossRefPubMedPubMedCentralGoogle Scholar
- 24.Cejas P, Lopez-Gomez M, Aguayo C, Madero R, de Castro Carpeno J, Belda-Iniesta C, Barriuso J, Moreno Garcia V, Larrauri J, Lopez R, Casado E, Gonzalez-Baron M, Feliu J. KRAS mutations in primary colorectal cancer tumors and related metastases: a potential role in prediction of lung metastasis. PloS one. 2009;4:e8199.CrossRefPubMedPubMedCentralGoogle Scholar
- 26.Cheng Y, Tian H. Current Development Status of MEK Inhibitors. Molecules. 2017;22.Google Scholar
- 29.Margonis GA, Sasaki K, Andreatos N, Kim Y, Merath K, Wagner D, Wilson A, Buettner S, Amini N, Antoniou E, Pawlik TM. KRAS Mutation Status Dictates Optimal Surgical Margin Width in Patients Undergoing Resection of Colorectal Liver Metastases. Annals of surgical oncology. 2017;24:264–271.CrossRefPubMedGoogle Scholar
- 30.Margonis GA, Sasaki K, Kim Y, Samaha M, Buettner S, Amini N, Antoniou E, Pawlik TM. Tumor Biology Rather Than Surgical Technique Dictates Prognosis in Colorectal Cancer Liver Metastases. Journal of gastrointestinal surgery : official journal of the Society for Surgery of the Alimentary Tract. 2016;20:1821–1829.CrossRefGoogle Scholar
- 31.Margonis GA, Buettner S, Andreatos N, Sasaki K, Ijzermans JNM, van Vugt JLA, Pawlik TM, Choti MA, Cameron JL, He J, Wolfgang CL, Weiss MJ. Anatomical Resections Improve Disease-free Survival in Patients With KRAS-mutated Colorectal Liver Metastases. Annals of surgery. 2017;266:641–649.CrossRefPubMedGoogle Scholar
- 32.Vauthey JN, Zimmitti G, Kopetz SE, Shindoh J, Chen SS, Andreou A, Curley SA, Aloia TA, Maru DM. RAS mutation status predicts survival and patterns of recurrence in patients undergoing hepatectomy for colorectal liver metastases. Annals of surgery. 2013;258:619–626; discussion 626-617.CrossRefPubMedGoogle Scholar
- 34.Petrowsky H, Sturm I, Graubitz O, Kooby DA, Staib-Sebler E, Gog C, Kohne CH, Hillebrand T, Daniel PT, Fong Y, Lorenz M. Relevance of Ki-67 antigen expression and K-ras mutation in colorectal liver metastases. European journal of surgical oncology : the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology. 2001;27:80–87.CrossRefGoogle Scholar
- 36.Brudvik KW, Mise Y, Chung MH, Chun YS, Kopetz SE, Passot G, Conrad C, Maru DM, Aloia TA, Vauthey JN. RAS Mutation Predicts Positive Resection Margins and Narrower Resection Margins in Patients Undergoing Resection of Colorectal Liver Metastases. Annals of surgical oncology. 2016;23:2635–2643.CrossRefPubMedPubMedCentralGoogle Scholar
- 39.Sasaki K, Margonis GA, Wilson A, Kim Y, Buettner S, Andreatos N, Gani F, Amini N, Spolverato G, Pawlik TM. Prognostic Implication of KRAS Status after Hepatectomy for Colorectal Liver Metastases Varies According to Primary Colorectal Tumor Location. Annals of surgical oncology. 2016;23:3736–3743.CrossRefPubMedGoogle Scholar
- 40.Sasaki K, Andreatos N, Margonis GA, He J, Weiss M, Johnston F, Wolfgang C, Antoniou E, Pikoulis E, Pawlik TM. The prognostic implications of primary colorectal tumor location on recurrence and overall survival in patients undergoing resection for colorectal liver metastasis. Journal of surgical oncology. 2016;114:803–809.CrossRefPubMedGoogle Scholar
- 41.Andreyev HJ, Norman AR, Cunningham D, Oates J, Dix BR, Iacopetta BJ, Young J, Walsh T, Ward R, Hawkins N, Beranek M, Jandik P, Benamouzig R, Jullian E, Laurent-Puig P, Olschwang S, Muller O, Hoffmann I, Rabes HM, Zietz C, Troungos C, Valavanis C, Yuen ST, Ho JW, Croke CT, O'Donoghue DP, Giaretti W, Rapallo A, Russo A, Bazan V, Tanaka M, Omura K, Azuma T, Ohkusa T, Fujimori T, Ono Y, Pauly M, Faber C, Glaesener R, de Goeij AF, Arends JW, Andersen SN, Lovig T, Breivik J, Gaudernack G, Clausen OP, De Angelis PD, Meling GI, Rognum TO, Smith R, Goh HS, Font A, Rosell R, Sun XF, Zhang H, Benhattar J, Losi L, Lee JQ, Wang ST, Clarke PA, Bell S, Quirke P, Bubb VJ, Piris J, Cruickshank NR, Morton D, Fox JC, Al-Mulla F, Lees N, Hall CN, Snary D, Wilkinson K, Dillon D, Costa J, Pricolo VE, Finkelstein SD, Thebo JS, Senagore AJ, Halter SA, Wadler S, Malik S, Krtolica K, Urosevic N. Kirsten ras mutations in patients with colorectal cancer: the 'RASCAL II' study. British journal of cancer. 2001;85:692–696.CrossRefPubMedPubMedCentralGoogle Scholar
- 42.Imamura Y, Morikawa T, Liao X, Lochhead P, Kuchiba A, Yamauchi M, Qian ZR, Nishihara R, Meyerhardt JA, Haigis KM, Fuchs CS, Ogino S. Specific mutations in KRAS codons 12 and 13, and patient prognosis in 1075 BRAF wild-type colorectal cancers. Clinical cancer research : an official journal of the American Association for Cancer Research. 2012;18:4753–4763.CrossRefGoogle Scholar
- 44.Ahronian LG, Sennott EM, Van Allen EM, Wagle N, Kwak EL, Faris JE, Godfrey JT, Nishimura K, Lynch KD, Mermel CH, Lockerman EL, Kalsy A, Gurski JM, Jr., Bahl S, Anderka K, Green LM, Lennon NJ, Huynh TG, Mino-Kenudson M, Getz G, Dias-Santagata D, Iafrate AJ, Engelman JA, Garraway LA, Corcoran RB. Clinical Acquired Resistance to RAF Inhibitor Combinations in BRAF-Mutant Colorectal Cancer through MAPK Pathway Alterations. Cancer discovery. 2015;5:358–367.CrossRefPubMedPubMedCentralGoogle Scholar
- 47.Souglakos J, Philips J, Wang R, Marwah S, Silver M, Tzardi M, Silver J, Ogino S, Hooshmand S, Kwak E, Freed E, Meyerhardt JA, Saridaki Z, Georgoulias V, Finkelstein D, Fuchs CS, Kulke MH, Shivdasani RA. Prognostic and predictive value of common mutations for treatment response and survival in patients with metastatic colorectal cancer. British journal of cancer. 2009;101:465–472.CrossRefPubMedPubMedCentralGoogle Scholar
- 48.Renfro LA, Goldberg RM, Grothey A, Sobrero A, Adams R, Seymour MT, Heinemann V, Schmoll HJ, Douillard JY, Hurwitz H, Fuchs CS, Diaz-Rubio E, Porschen R, Tournigand C, Chibaudel B, Hoff PM, Kabbinavar FF, Falcone A, Tebbutt NC, Punt CJA, Hecht JR, Souglakos J, Bokemeyer C, Van Cutsem E, Saltz L, de Gramont A, Sargent DJ, Program ACT. Clinical Calculator for Early Mortality in Metastatic Colorectal Cancer: An Analysis of Patients From 28 Clinical Trials in the Aide et Recherche en Cancerologie Digestive Database. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2017;35:1929–1937.CrossRefGoogle Scholar
- 50.Karagkounis G, Torbenson MS, Daniel HD, Azad NS, Diaz LA, Jr., Donehower RC, Hirose K, Ahuja N, Pawlik TM, Choti MA. Incidence and prognostic impact of KRAS and BRAF mutation in patients undergoing liver surgery for colorectal metastases. Cancer. 2013;119:4137–4144.Google Scholar
- 51.Bokemeyer C, Van Cutsem E, Rougier P, Ciardiello F, Heeger S, Schlichting M, Celik I, Kohne CH. Addition of cetuximab to chemotherapy as first-line treatment for KRAS wild-type metastatic colorectal cancer: pooled analysis of the CRYSTAL and OPUS randomised clinical trials. European journal of cancer. 2012;48:1466–1475.CrossRefPubMedGoogle Scholar
- 52.Schirripa M, Bergamo F, Cremolini C, Casagrande M, Lonardi S, Aprile G, Yang D, Marmorino F, Pasquini G, Sensi E, Lupi C, De Maglio G, Borrelli N, Pizzolitto S, Fasola G, Bertorelle R, Rugge M, Fontanini G, Zagonel V, Loupakis F, Falcone A. BRAF and RAS mutations as prognostic factors in metastatic colorectal cancer patients undergoing liver resection. British journal of cancer. 2015;112:1921–1928.CrossRefPubMedPubMedCentralGoogle Scholar
- 53.Margonis GA, Buettner S, Andreatos N, Wagner D, Sasaki K, Barbon C, Beer A, Kamphues C, Løes IM, He J, Pawlik TM, Kaczirek K, Poultsides G, Lønning PE, Cameron JL, Mischinger HJ, Aucejo FN, Kreis ME, Wolfgang CL WM. Prognostic factors change over time following hepatectomy for Colorectal Liver Metastases : A multi-institutional , international analysis of 1099 patients. Ann Surg. 2017.Google Scholar
- 55.Chun YS, Passot G, Yamashita S, Nusrat M, Katsonis P, Loree JM, Conrad C, Tzeng CD, Xiao L, Aloia TA, Eng C, Kopetz SE, Lichtarge O, Vauthey JN. Deleterious Effect of RAS and Evolutionary High-risk TP53 Double Mutation in Colorectal Liver Metastases. Annals of surgery. 2017.Google Scholar
- 56.Ivanecz A, Kavalar R, Palfy M, Pivec V, Sremec M, Horvat M, Potrc S. Can we improve the clinical risk score? The prognostic value of p53, Ki-67 and thymidylate synthase in patients undergoing radical resection of colorectal liver metastases. HPB : the official journal of the International Hepato Pancreato Biliary Association. 2014;16:235–242.CrossRefGoogle Scholar
- 57.Nitti D, Belluco C, Montesco MC, Bertorelle R, Da Pian PP, Fassina A, Ninfo V, Chieco-Bianchi L, Lise M. Nuclear p53 protein expression in resected hepatic metastases from colorectal cancer: an independent prognostic factor of survival. European journal of cancer. 1998;34:851–855.CrossRefPubMedGoogle Scholar
- 58.Pilat N, Grunberger T, Langle F, Mittlbock M, Perisanidis B, Kappel S, Wolf B, Starlinger P, Kuhrer I, Muhlbacher F, Kandioler D. Assessing the TP53 marker type in patients treated with or without neoadjuvant chemotherapy for resectable colorectal liver metastases: a p53 Research Group study. European journal of surgical oncology : the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology. 2015;41:683–689.CrossRefGoogle Scholar
- 60.Yang Y, Forslund A, Remotti H, Lonnroth C, Andersson M, Brevinge H, Svanberg E, Lindner P, Hafstrom L, Naredi P, Lundholm K. P53 mutations in primary tumors and subsequent liver metastases are related to survival in patients with colorectal carcinoma who undergo liver resection. Cancer. 2001;91:727–736.CrossRefPubMedGoogle Scholar
- 62.Yamashita S, Chun YS, Kopetz SE, Maru D, Conrad C, Aloia TA, Vauthey JN. APC and PIK3CA Mutational Cooperativity Predicts Pathologic Response and Survival in Patients Undergoing Resection for Colorectal Liver Metastases. Annals of surgery. 2017.Google Scholar
- 63.Samuels Y, Diaz LA, Jr., Schmidt-Kittler O, Cummins JM, Delong L, Cheong I, Rago C, Huso DL, Lengauer C, Kinzler KW, Vogelstein B, Velculescu VE. Mutant PIK3CA promotes cell growth and invasion of human cancer cells. Cancer cell. 2005;7:561–573.Google Scholar
- 65.Sartore-Bianchi A, Martini M, Molinari F, Veronese S, Nichelatti M, Artale S, Di Nicolantonio F, Saletti P, De Dosso S, Mazzucchelli L, Frattini M, Siena S, Bardelli A. PIK3CA mutations in colorectal cancer are associated with clinical resistance to EGFR-targeted monoclonal antibodies. Cancer research. 2009;69:1851–1857.CrossRefPubMedGoogle Scholar
- 66.Prenen H, De Schutter J, Jacobs B, De Roock W, Biesmans B, Claes B, Lambrechts D, Van Cutsem E, Tejpar S. PIK3CA mutations are not a major determinant of resistance to the epidermal growth factor receptor inhibitor cetuximab in metastatic colorectal cancer. Clinical cancer research : an official journal of the American Association for Cancer Research. 2009;15:3184–3188.Google Scholar
- 67.Loes IM, Immervoll H, Sorbye H, Angelsen JH, Horn A, Knappskog S, Lonning PE. Impact of KRAS, BRAF, PIK3CA, TP53 status and intraindividual mutation heterogeneity on outcome after liver resection for colorectal cancer metastases. International journal of cancer. 2016;139:647–656.CrossRefPubMedPubMedCentralGoogle Scholar
- 72.Goldstein J, Tran B, Ensor J, Gibbs P, Wong HL, Wong SF, Vilar E, Tie J, Broaddus R, Kopetz S, Desai J, Overman MJ. Multicenter retrospective analysis of metastatic colorectal cancer (CRC) with high-level microsatellite instability (MSI-H). Annals of oncology : official journal of the European Society for Medical Oncology. 2014;25:1032–1038.Google Scholar
- 74.Muller CI, Schulmann K, Reinacher-Schick A, Andre N, Arnold D, Tannapfel A, Arkenau H, Hahn SA, Schmoll SH, Porschen R, Schmiegel W, Graeven U, Group AIOCS. Predictive and prognostic value of microsatellite instability in patients with advanced colorectal cancer treated with a fluoropyrimidine and oxaliplatin containing first-line chemotherapy. A report of the AIO Colorectal Study Group. International journal of colorectal disease. 2008;23:1033–1039.CrossRefPubMedGoogle Scholar
- 75.Kochhar R, Halling KC, McDonnell S, Schaid DJ, French AJ, O'Connell MJ, Nagorney DM, Thibodeau SN. Allelic imbalance and microsatellite instability in resected Duke's D colorectal cancer. Diagnostic molecular pathology : the American journal of surgical pathology, part B. 1997;6:78–84.CrossRefGoogle Scholar
- 76.Haddad R, Ogilvie RT, Croitoru M, Muniz V, Gryfe R, Pollet A, Shanmugathasan P, Fitzgerald T, Law CH, Hanna SS, Jothy S, Redston M, Gallinger S, Smith AJ. Microsatellite instability as a prognostic factor in resected colorectal cancer liver metastases. Annals of surgical oncology. 2004;11:977–982.CrossRefPubMedGoogle Scholar
- 80.Rekhtman N, Ang DC, Riely GJ, Ladanyi M, Moreira AL. KRAS mutations are associated with solid growth pattern and tumor-infiltrating leukocytes in lung adenocarcinoma. Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc. 2013;26:1307–1319.CrossRefGoogle Scholar
- 81.Shindoh J, Makuuchi M, Matsuyama Y, Mise Y, Arita J, Sakamoto Y, Hasegawa K, Kokudo N. Complete removal of the tumor-bearing portal territory decreases local tumor recurrence and improves disease-specific survival of patients with hepatocellular carcinoma. Journal of hepatology. 2016;64:594–600.CrossRefPubMedGoogle Scholar
- 82.Zorzi D, Mullen JT, Abdalla EK, Pawlik TM, Andres A, Muratore A, Curley SA, Mentha G, Capussotti L, Vauthey JN. Comparison between hepatic wedge resection and anatomic resection for colorectal liver metastases. Journal of gastrointestinal surgery : official journal of the Society for Surgery of the Alimentary Tract. 2006;10:86–94.CrossRefGoogle Scholar
- 83.Jones JC, Renfro LA, Al-Shamsi HO, Schrock AB, Rankin A, Zhang BY, Kasi PM, Voss JS, Leal AD, Sun J, Ross J, Ali SM, Hubbard JM, Kipp BR, McWilliams RR, Kopetz S, Wolff RA, Grothey A. Non-V600 BRAF Mutations Define a Clinically Distinct Molecular Subtype of Metastatic Colorectal Cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2017;35:2624–2630.CrossRefGoogle Scholar
- 88.Le DT, Uram JN, Wang H, Bartlett BR, Kemberling H, Eyring AD, Skora AD, Luber BS, Azad NS, Laheru D, Biedrzycki B, Donehower RC, Zaheer A, Fisher GA, Crocenzi TS, Lee JJ, Duffy SM, Goldberg RM, de la Chapelle A, Koshiji M, Bhaijee F, Huebner T, Hruban RH, Wood LD, Cuka N, Pardoll DM, Papadopoulos N, Kinzler KW, Zhou S, Cornish TC, Taube JM, Anders RA, Eshleman JR, Vogelstein B, Diaz LA, Jr. PD-1 Blockade in Tumors with Mismatch-Repair Deficiency. The New England journal of medicine. 2015;372:2509–2520.CrossRefPubMedPubMedCentralGoogle Scholar