ABSTRACT
Several trials have demonstrated that bevacizumab in combination with standard chemotherapy improved response rates, progression free survival (PFS) and overall survival (OS) when compared with chemotherapy alone in the treatment of patients with mCRC [38-42]. In hopes of improving outcomes in patients with mCRC, researchers have attempted dual inhibition of both the VEGF and EGFR pathways by combining bevacizumab and chemotherapy with either cetuximab or panitumumab [43-45]. Unfortunately, these trials did not demonstrate any clinical benefit with the addition of an anti-EGFR moAb and may have been a detriment to survival (See Table 7.2) [43, 44]. In contrast to result with moAbs, phase 2 trials using the EGFR TKIs, erlotinib and gefitinib, have been disappointing as monotherapy did not demonstrate activity when tested on patients with previously treated mCRC [46, 47]. There is some phase 2 data suggesting that these agents may be useful when combined with chemotherapy [48-50], however in the absence of completed phase 3 trials demonstrating clinical benefit, these agents remain experimental in the treatment of patients with mCRC. 7.4 Determinants of Sensitivity and Resistance
The benefit of cetuximab and panitumumab in the treatment of patients with mCRC has been clearly demonstrated, albeit with only modest objective response rates of approximately 10% and improvements of survival of 3-4 months [31, 35]. These response and survival rates improve, by combining these anti-EGFR moAbs with multi-agent chemotherapy and using these regimens in treatment naïve patients with mCRC [34]. To further improve the treatment of patients with mCRC it is imperative to identify predictive markers of response or resistance to cetuximab and panitumumab. These markers will aide in identifying patients who may benefit from treatment, while avoiding the unnecessary toxicities and costs associated with these drugs in those who may not. A predictive marker is not necessarily the same as a prognostic marker. A predictive marker gives information about the effect of a therapeutic intervention, whilst a prognostic marker provides information about a patients overall cancer outcome, regardless
of therapy. An ideal predictive marker is simple, reproducible, correlates with a clinical endpoint, and is available prior to initiation of treatment. Both clinical and cellular alterations have emerged as potential predictive markers for anti-EGFR moAb therapies. However, as with most predictive tests, none have been ideal at predicting response or resistance to anti-EGFR moAbs. This is, in part, due to the complexities in identifying appropriate markers for testing and the scoring of these markers. EGFR inhibitor induced skin rash is a clinical feature that has been evaluated as a potential predictive marker. The cellular features that alter DNA, microRNA, and/or protein in the EGFR pathway from colorectal cancers have also been evaluated for predictive potential. 7.4.1 Clinical Features
The toxicity profile of the anti-EGFR moAbs excludes several of the severe side effects commonly associated with cytotoxic chemotherapy, such as hematological suppression. However, the use of these agents have been hampered by a constellation of dermatological toxicities experienced by more than 90% of patients, notably, papulopustular rash, xerosis, pruritis, and paronychia [51]. The patho-physiology of the cetuximab-induced rash remains poorly understood, but is likely due to the predicted downstream perturbations of EGFR signaling that have been demonstrated in studies of skin biopsies, such as phosphorylated EGFR, phosphorylated AKT, Ki67, and p27 in patients treated with anti-EGFR moAbs [52-54]. Though rarely life threatening, these skin effects impact quality of life, increase risk for infection, and may lead to dose modification or treatment discontinuation, thus potentially affecting the overall clinical benefits of therapy. A significant positive correlation between cutaneous toxicity and rates of response, PFS, and OS has been noted in all trials of cetuximab [34], or panitumumab [35] in patients with advanced colorectal cancer. Based on the association of cetuximab and rash, the EVEREST trial was launched [55]. Patients with irinotecan-refractory EGFR-expressing CRC were randomized to either a standard dose of cetuximab or the dose escalation arm, which increased the cetuximab dose up to 500 mg/m2 if the patient did not experience more than Grade 1 rash by week 3. Preliminary results
from this trial suggests that cetuximab dose escalation is safe and increases response rates in patients without a rash or those who develop only a mild skin reaction within 3 weeks of starting therapy. The response rate among patients with less than or equal to a grade 1 skin toxicity who were randomized to dose escalation was 30% versus 13% in the control group who only continued standard dose therapy [55]. This trend was not statistically significant and requires more mature data from a larger sample population. Using rash as an early clinical marker of benefit from anti-EGFR moAbs does have some limitations. Rash has occurred in patients without any apparent benefit from anti-EGFR moAbs, and conversely clinical benefit has occurred in patients without rash. Some suspect that rash may be a surrogate marker for a genetic profile that predisposes individuals to rash and also modulate the response to anti-EGFR moAbs. Also, rash could merely be a prognostic marker of intact immunity or merely the result of treatment bias; as patients who have intact immune systems or are treated long enough to develop a rash may have prolonged survival [56]. Although, a clear association between signaling inhibition in skin and antitumor response has not been found, rash seems to correlate with drug exposure, concentration, and likely reflects effectiveness.
