**3. Systemic cytotoxic chemotherapy**

Several nonrandomized and phase I, II, and III clinical trials have been conducted to investigate the role of systemic cytotoxic chemotherapy (monotherapy or combination therapy) for the management of advanced inoperable HCC.

#### **3.1. Monotherapy (single-agent) systemic chemotherapy**

Several single-agent systemic chemotherapies have been tested in patients with advanced HCC, such as: doxorubicin, pegylated liposomal doxorubicin (PLD), epirubicin, mitoxantrone, 5-fluorouracil (5-FU), etoposide, capecitabine, gemcitabine, irinotecan, and thalidomide.

#### *3.1.1. Doxorubicin*

Single-agent doxorubicin is the most frequently investigated systemic chemotherapeutic agent in patients with locally advanced unresectable HCC [33].

In 1975, Olweny et al. [34] (phase II clinical trial) studied the role of doxorubicin (75 mg/m2 intravenously once every 3 weeks) in 14 patients with primary advanced inoperable HCC. Eleven patients (78.5%) achieved objective responses (78.5%). However, successive studies (from 1977 to 2005) failed to validate Olweny et al. [34] study and rather exhibited that the actual objective response rate with single-agent doxorubicin (dose: 75 mg/m2 ) was rough‐ ly equal to or less than 20% [35-40]. Additional large-sized subsequent randomized trials employing lower doses of single-agent doxorubicin (dose: equals to or less than 60 mg/m2 per schedule) were shown to yield even lower objective response rates ranging from 4% to 10.1% [41-42].

In 1988, Lai et al. [39] (prospective randomized trial) studied the efficacy of doxorubicin (60-75 mg/m2 ) versus the best supportive care (no chemotherapy) in 60 and 46 patients, respectively. The doxorubicin-treated group achieved higher statistically significant median OS than the no chemotherapy group (10.6 vs. 7.6 weeks; *P* = 0.036). However, life-threatening toxicities (cardiotoxicity and septicemia) occurred in the doxorubicin-treated group (25%). The study concluded that despite the minimal survival advantages of doxorubicin, it was associated with serious complications and should not be recommended for the management of inoperable HCC.

In 2007, Gish et al. [42] (phase III randomized controlled trial) examined the efficacy of doxorubicin versus nolatrexed in 445 patients. The doxorubicin-treated group achieved a higher statistically significant OS than nolatrexed-treated group (32.3 vs. 22.3 weeks; *P* = 0.0068). The objective response rates for doxorubicin-treated and nolatrexed-treated groups were 4% and 1.4%, respectively. The most frequently observed toxicities for doxorubicintreated and nolatrexed-treated groups were alopecia and grade 3/4 (thrombocytopenia, vomiting, diarrhea, and stomatitis), respectively.

In conclusion, single-agent doxorubicin can be effective in 20% of patients; however, OS advantages are uncertain. Moreover, its cardiotoxicity is a major limiting adverse event. Combination therapy with other systemic cytotoxic chemotherapeutics and novel molecularly targeted therapies are in progress.

#### *3.1.2. Pegylated liposomal doxorubicin (PLD)*

The efficacy of single-agent PLD has been studied in a pilot study [43] and two phase II trials [44, 45] as an initial therapy in patients with advanced inoperable HCC. The research outcomes were discouraging. Combination chemotherapeutic remedies containing PLD are elaborated below.

#### *3.1.3. Epirubicin and mitoxantrone*

In comparison with doxorubicin, previous retrospective studies and phase II trials demon‐ strated that single-agent epirubicin [46, 47] and mitoxantrone [48, 49] share relatively compa‐ rable antineoplastic activity as well as relatively equal or slightly higher objective response rates (epirubicin, range: 9.1%-23%; mitoxantrone, range: 23.7%-27.2%). Cardiotoxicity is a major limiting adverse event. Both chemotherapeutics are not commonly used.

### *3.1.4. 5-Fluorouracil (5-FU)*

In 1975, Olweny et al. [34] (phase II clinical trial) studied the role of doxorubicin (75 mg/m2 intravenously once every 3 weeks) in 14 patients with primary advanced inoperable HCC. Eleven patients (78.5%) achieved objective responses (78.5%). However, successive studies (from 1977 to 2005) failed to validate Olweny et al. [34] study and rather exhibited that the

ly equal to or less than 20% [35-40]. Additional large-sized subsequent randomized trials employing lower doses of single-agent doxorubicin (dose: equals to or less than 60 mg/m2 per schedule) were shown to yield even lower objective response rates ranging from 4% to

In 1988, Lai et al. [39] (prospective randomized trial) studied the efficacy of doxorubicin (60-75

In 2007, Gish et al. [42] (phase III randomized controlled trial) examined the efficacy of doxorubicin versus nolatrexed in 445 patients. The doxorubicin-treated group achieved a higher statistically significant OS than nolatrexed-treated group (32.3 vs. 22.3 weeks; *P* = 0.0068). The objective response rates for doxorubicin-treated and nolatrexed-treated groups were 4% and 1.4%, respectively. The most frequently observed toxicities for doxorubicintreated and nolatrexed-treated groups were alopecia and grade 3/4 (thrombocytopenia,

In conclusion, single-agent doxorubicin can be effective in 20% of patients; however, OS advantages are uncertain. Moreover, its cardiotoxicity is a major limiting adverse event. Combination therapy with other systemic cytotoxic chemotherapeutics and novel molecularly

The efficacy of single-agent PLD has been studied in a pilot study [43] and two phase II trials [44, 45] as an initial therapy in patients with advanced inoperable HCC. The research outcomes were discouraging. Combination chemotherapeutic remedies containing PLD are elaborated

In comparison with doxorubicin, previous retrospective studies and phase II trials demon‐ strated that single-agent epirubicin [46, 47] and mitoxantrone [48, 49] share relatively compa‐ rable antineoplastic activity as well as relatively equal or slightly higher objective response rates (epirubicin, range: 9.1%-23%; mitoxantrone, range: 23.7%-27.2%). Cardiotoxicity is a

major limiting adverse event. Both chemotherapeutics are not commonly used.

vomiting, diarrhea, and stomatitis), respectively.

targeted therapies are in progress.

*3.1.3. Epirubicin and mitoxantrone*

*3.1.2. Pegylated liposomal doxorubicin (PLD)*

) versus the best supportive care (no chemotherapy) in 60 and 46 patients, respectively. The doxorubicin-treated group achieved higher statistically significant median OS than the no chemotherapy group (10.6 vs. 7.6 weeks; *P* = 0.036). However, life-threatening toxicities (cardiotoxicity and septicemia) occurred in the doxorubicin-treated group (25%). The study concluded that despite the minimal survival advantages of doxorubicin, it was associated with serious complications and should not be recommended for the management of inoperable

) was rough‐

actual objective response rate with single-agent doxorubicin (dose: 75 mg/m2

10.1% [41-42].

110 Recent Advances in Liver Diseases and Surgery

mg/m2

HCC.

below.

In one prospective randomized controlled trial by Choi et al. [37], there were higher objective response rates and median OS in HCC patients receiving doxorubicin versus those patients receiving 5-fluorouracil-containg quadruple therapy (5-fluorouracil, methotrexate, cyclophos‐ phamide, and vincristin) therapy (24% vs. 0%, respectively; 14.4 vs. 6.5 weeks, respectively).

In 1995, Porta et al. [50] (preliminary results of a phase II study) explored the role of 5-FU (370 mg/m2 ) plus racemic leucovorin (200 mg/m2 ) for 5 successive days in 25 patients with advanced inoperable HCC. The regimen cycle was continual every 28 days until disease progression took place. Seven objective responses (28%) were achieved as follows: 6 partial (24%) and 1 complete (4%) responses. Only 5 patients (20%) displayed stable disease, whereas 13 patients exhibited disease progression. Regimen-related adverse events were mild and no grade 4 toxicity occurred. Specifically, 1 patient (4%) experienced grade 1 skin toxicity, 2 patients (8%) grade 3 granulocytopenia, 7 patients (28%) grade 2 nausea, 10 patients (40%) grade 2 diarrhea, and 11 patients (44%) grade 2/3 mucositis. The study concluded that (5-FU plus racemic leucovorin) chemotherapeutic schedule could provide objective responses in patients with advanced unresectable HCCs, which are frequently regarded as chemoresistant neoplasms.

In 1995, Tetef et al. [51] (phase II trial) examined the role of 5-FU (250-450 mg/m2 /day for 5 days by means of an intravenous [IV] bolus) in combination with calcium leucovorin (500 mg/m2 /day for 5 days by means of continuous IV infusion) in 15 patients with advanced unresectable HCC. The regimen was given on a 28-day schedule. Overall, 8 (53%), 6 (40%), and 1 (7%) patients experienced stable disease, disease progression, and partial response, respectively. The median duration of stable disease was 5.7 months, whereas the median TTP was 2.7 months and the partial response persisted only for 2.4 months. Overall, the median OS was roughly 4 months. Regarding regimen-related adverse events, only 9% and 10% of chemotherapeutic schedules were impacted negatively by grade 3/4 hematological toxicity and grade 3/4 gastrointestinal toxicity, respectively. The study concluded that (5-FU plus calcium leucovorin) chemotherapeutic schedule is ineffective highlighting the chemoresistant characteristic of HCC to the modulated 5-FU.

In conclusion, objective response rates with single-agent 5-FU have been frequently low despite the addition of modulating agents such as leucovorin. Advantageously, despite the widespread hepatic metabolism, satisfactory doses of 5-FU can be often administered in HCC patients with hepatic insufficiency or jaundice.

#### *3.1.5. Etoposide*

An early prospective randomized controlled trial demonstrated higher ORR (however no survival advantages) when single-agent doxorubicin was contrasted to single-agent etoposide (28% vs. 18%, respectively)[52].

Further trials are underway to test its true efficacy both singly and in combination with other drugs in the management of HCC.

#### *3.1.6. Capecitabine*

In 2004, Patt et al. [53] (retrospective analysis) studied the role of single-agent oral capecitabine (1000 mg/m2 twice daily for 2 weeks; treatment was repeated every 3 weeks) in 37 patients with advanced inoperable HCC. Of the 37 patients, 22 patients had not received any previous treatment. Objective responses were attained in 9 patients (24.3%), comprising 1 complete response. The median OS was 10.1 months. Grade 3 thrombocytopenia happened in 3 patients. The study concluded that capecitabine is well tolerated and offers only minimal antitumor activities against HCC.

In 2013, Brandi et al. [54] (single-center phase II study) examined the role of single-agent metronomic capecitabine (500 mg twice daily) in 90 patients with advanced HCC. The patients were divided into two groups. The first group consisted of 59 patients who had received no prior therapy. Three objective responses (1 partial and 2 complete) were attained whereas 30 patients experienced stable disease. The median PFS and OS were 6.03 and 14.47 months, respectively. The second group consisted of 31 patients who received prior therapy with sorafenib. No objective responses (neither partial nor complete) were attained whereas 10 patients experienced stable disease. The median PFS and OS were 3.27 and 9.77 months, respectively. The first group (capecitabine-treated) was matched to untreated HCC patients from the Italian Liver Cancer group. The capecitabine-treated group achieved a higher statistically significant median OS than the matched untreated patients (15.6 months vs. 8.0 months; *P* = 0.043). The study concluded that metronomic capecitabine seems to offer anti‐ neoplastic activities in therapy-naive and sorafenib-treated patients.

The superiority of single-agent sorafenib over capecitabine was confirmed in a single-center, open-label, phase II trial by Abdel-Rahman et al. [55]. The study enrolled 52 treatment-naive HCC patients who were randomized to get administered sorafenib (400 mg twice daily) or capecitabine (100 mg mg/m2 twice daily). In comparison with the capecitabine-treated group, the sorafenib-treated group achieved higher statistically significant median PFS (6 months vs. 4 months; *P* < 0.005) and OS (7.05 vs. 5.07 months; *P* < 0.016). Four objective responses (3 partial and 1 complete) were achieved in sorafenib-treated group; only 1 partial response was achieved in capecitabine-treated group. The most commonly observed toxicities in sorafenibtreated and capecitabine-treated groups were hand-foot skin reaction and hyperbilirubinemia, respectively. The study concluded that (1) sorafenib is superior to capecitabine in patients with HCC and (2) capecitabine should not be employed as a single-agent therapy; instead, combi‐ nation regimens with sorafenib should be attempted.

In conclusion, the DFS and OS advantages of single-agent fluoropyrimidines (5-FU and capecitabine) are uncertain, partly due to inconsistent study participants (treatment naive and previously treated). Combination regimens with other chemotherapeutic agents should be examined in phase II/III clinical trials.

#### *3.1.7. Gemcitabine*

Single-agent gemcitabine chemotherapy has showed varied modest results in 3 phase II clinical trials [56-58].

In 2000, Yang et al. [56] studied the role of gemcitabine (intravenous 1250 mg/m2 once weekly for 3 weeks followed by a 1-week rest) in 28 chemotherapy-naive patients with inoperable, nonembolizable, locally advanced or metastatic HCC. All study patients received 6 cycles of gemcitabine, as follows: 1250 mg/m2 once weekly for 3 weeks followed by a 1-week rest. Partial response was attained in 5 of 28 patients (overall response rate: 17.8%). Stable disease was attained in 7 patients (25%). Disease progression occurred in 16 patients (57.2%). The median OS in all the 28 patients and those 5 patients who had partial response was 18.7 and 34.7 weeks, respectively. The median TTP was roughly 12 weeks. Grade 3/4 adverse events mainly comprised equally thrombocytopenia and leucopenia (10.7%) as well as equally anemia and hepatotoxicity (14.3%).

In 2001, Kubicka et al. [57] studied the role of gemcitabine in 20 patients with advanced unresectable HCC. The median number of gemcitabine administration was 7.6 (range: 3-21). The overall response rate was attained in 1 patient (5%), and gemcitabine did not ameliorate the cancer-related symptoms. Grade 3/4 thrombocytopenia was the most commonly observed adverse event (30%).

In 2002, Fuchs et al. [58] studied the role of gemcitabine (intravenous 1000 mg/m2 once weekly for 3 weeks followed by 1 resting week) in 30 patients with advanced unresectable metastatic HCC. The enrolled patients had received at least one prior modality of systemic therapy in the past. The median number of gemcitabine administration was 2 (range: 1-8). Neither complete nor partial responses were attained. Only 9 patients (30%) attained stable disease (median interval: 7.4 months). The median OS was 6.9 months, whereas the overall 1-year survival was 40%. One patient (3%) suffered grade 3 thrombocytopenia whereas another one patient (3%) suffered hemolytic-uremic syndrome. Additionally, 2 patients (7%) developed grade 4 neutropenia.

In conclusion, although gemcitabine is largely well tolerated, phase II clinical trials of gemci‐ tabine exhibited minimal effects in patients with advanced unresectable HCC and therefore is not recommended. Gemcitabine-based combination therapies are interesting therapeutic targets.

#### *3.1.8. Thalidomide*

*3.1.6. Capecitabine*

112 Recent Advances in Liver Diseases and Surgery

activities against HCC.

(1000 mg/m2

In 2004, Patt et al. [53] (retrospective analysis) studied the role of single-agent oral capecitabine

with advanced inoperable HCC. Of the 37 patients, 22 patients had not received any previous treatment. Objective responses were attained in 9 patients (24.3%), comprising 1 complete response. The median OS was 10.1 months. Grade 3 thrombocytopenia happened in 3 patients. The study concluded that capecitabine is well tolerated and offers only minimal antitumor

In 2013, Brandi et al. [54] (single-center phase II study) examined the role of single-agent metronomic capecitabine (500 mg twice daily) in 90 patients with advanced HCC. The patients were divided into two groups. The first group consisted of 59 patients who had received no prior therapy. Three objective responses (1 partial and 2 complete) were attained whereas 30 patients experienced stable disease. The median PFS and OS were 6.03 and 14.47 months, respectively. The second group consisted of 31 patients who received prior therapy with sorafenib. No objective responses (neither partial nor complete) were attained whereas 10 patients experienced stable disease. The median PFS and OS were 3.27 and 9.77 months, respectively. The first group (capecitabine-treated) was matched to untreated HCC patients from the Italian Liver Cancer group. The capecitabine-treated group achieved a higher statistically significant median OS than the matched untreated patients (15.6 months vs. 8.0 months; *P* = 0.043). The study concluded that metronomic capecitabine seems to offer anti‐

The superiority of single-agent sorafenib over capecitabine was confirmed in a single-center, open-label, phase II trial by Abdel-Rahman et al. [55]. The study enrolled 52 treatment-naive HCC patients who were randomized to get administered sorafenib (400 mg twice daily) or capecitabine (100 mg mg/m2 twice daily). In comparison with the capecitabine-treated group, the sorafenib-treated group achieved higher statistically significant median PFS (6 months vs. 4 months; *P* < 0.005) and OS (7.05 vs. 5.07 months; *P* < 0.016). Four objective responses (3 partial and 1 complete) were achieved in sorafenib-treated group; only 1 partial response was achieved in capecitabine-treated group. The most commonly observed toxicities in sorafenibtreated and capecitabine-treated groups were hand-foot skin reaction and hyperbilirubinemia, respectively. The study concluded that (1) sorafenib is superior to capecitabine in patients with HCC and (2) capecitabine should not be employed as a single-agent therapy; instead, combi‐

In conclusion, the DFS and OS advantages of single-agent fluoropyrimidines (5-FU and capecitabine) are uncertain, partly due to inconsistent study participants (treatment naive and previously treated). Combination regimens with other chemotherapeutic agents should be

Single-agent gemcitabine chemotherapy has showed varied modest results in 3 phase II clinical

neoplastic activities in therapy-naive and sorafenib-treated patients.

nation regimens with sorafenib should be attempted.

examined in phase II/III clinical trials.

*3.1.7. Gemcitabine*

trials [56-58].

twice daily for 2 weeks; treatment was repeated every 3 weeks) in 37 patients

Single-agent thalidomide chemotherapy has been investigated in 3 early phase II clinical trials -61]. Thalidomide showed lower rates of antineoplastic effects; however, disease stabilization was achieved in up to 33% of patients.

In 2003, Hsu et al. studied the role of low-dose thalidomide (starting dose of 200 mg per day; the dose was gradually upgraded in 100-mg phases up to maximum tolerated dose or 600 mg per day) in 68 patients with advanced unresectable HCC. Four patients (6.3%) attained chemotherapy responses (1 complete and 3 partial), and their AFP levels fell greatly. Moreover, an additional 6 patients experienced more than 50% reduction in their AFP levels post treatment with thalidomide. In total, 10 patients achieved objective response to thalidomide with a median OS of 62.4 weeks (range: 31.2-93.6). For all patients, the median OS was 18.7 weeks, whereas the overall 1-year survival was 27.6%. Only 6 and none patients developed grade 3 and grade 4 thalidomide-related adverse events, respectively.

In 2005, Lin et al. studied the role of thalidomide (starting dose of 200 mg per day; the dose was gradually upgraded in 100-mg phases up to maximum tolerated dose or 800 mg per day) in 27 patients with advanced unresectable HCC. The median daily dose was 300 mg. Only 1 patient achieved near-complete drug response (expressed as reduced AFP level) as well as partial radiological response on computed tomography (CT) imaging. Stable disease of 16 week interval was attained in 2 patients. The median DFS was 6 weeks, whereas the overall OS was 17.6 weeks. Fatigue (81%) and somnolence (62%) were the two most frequent thali‐ domide-related adverse events. Three patients suffered grade 4 hyperbilirubinemia.

In 2005, Patt et al. [61] studied the role of thalidomide (starting dose of 200 mg per day; the dose was gradually upgraded from 400 mg during the first week to 1000 mg during the fifth week) in 37 patients with advanced unresectable HCC. Overall, 1 (5%), 1 (5%), and 10 (31.3%) patients attained partial response, minor response, and stable disease, respectively. Twenty patients (62.5%) experienced disease progression. The overall OS was roughly 6.8 months. The most frequently observed drug-related adverse events were grade 2/3/4 somnolence in 65% whereas grade 3/4 reactions occurred in 20% of patients.

In conclusion, with gradual dose escalation, thalidomide exhibited well-tolerated toxicity profile. While thalidomide demonstrated lower response rates, it offered disease stabilization in one-third of patients. Future studies should be targeted toward exploring different thali‐ domide analogs and doses as well as trial of combination therapy with other systemic management modalities. As of now, thalidomide use in the management of advanced HCC is not recommended.

#### *3.1.9. Irinotecan*

Single-agent irinotecan chemotherapy has been investigated in two phase II clinical trials for the management of patients with advanced unresectable HCC [62,.

In 2001, O'Reilly et al. (phase II) studied the role of irinotecan (starting dose of 125 mg once weekly for 4 weeks followed by a 2-week rest) in 14 patients with advanced unresectable HCC. The median number of irinotecan cycle administration was 1 (range: 1-6). Partial response was attained in only 1 patient (7%), which lasted for 7 weeks. Transient stable disease was attained in 1 patient (7%). Disease progression occurred in all the 12 remaining patients (86%). Signif‐ icant irinotecan-related adverse events were noted, mainly nausea, vomiting, diarrhea, fatigue, and neutropenia.

In 2006, Boige et al. (multicenter phase II study) studied the role of irinotecan (dose was adjusted according to total bilirubin level) in 29 patients with advanced unresectable HCC. In total, 0, 1, and 12 patients experienced objective response, minor response, and disease stabilization, respectively. Median TTP was 3.1 months whereas the OS was 7.4 months. Grade 3/4 toxicities primarily compromised diarrhea (17%), anemia (24%), and neutropenia (47%).

In conclusion, irinotecan had considerable drug-related toxicities (adverse events) and very minimal antitumor effects against advanced unresectable HCC. Single-agent irinotecan chemotherapy is not recommended.

#### **3.2. Combination systemic cytotoxic chemotherapy**

weeks, whereas the overall 1-year survival was 27.6%. Only 6 and none patients developed

In 2005, Lin et al. studied the role of thalidomide (starting dose of 200 mg per day; the dose was gradually upgraded in 100-mg phases up to maximum tolerated dose or 800 mg per day) in 27 patients with advanced unresectable HCC. The median daily dose was 300 mg. Only 1 patient achieved near-complete drug response (expressed as reduced AFP level) as well as partial radiological response on computed tomography (CT) imaging. Stable disease of 16 week interval was attained in 2 patients. The median DFS was 6 weeks, whereas the overall OS was 17.6 weeks. Fatigue (81%) and somnolence (62%) were the two most frequent thali‐

domide-related adverse events. Three patients suffered grade 4 hyperbilirubinemia.

In 2005, Patt et al. [61] studied the role of thalidomide (starting dose of 200 mg per day; the dose was gradually upgraded from 400 mg during the first week to 1000 mg during the fifth week) in 37 patients with advanced unresectable HCC. Overall, 1 (5%), 1 (5%), and 10 (31.3%) patients attained partial response, minor response, and stable disease, respectively. Twenty patients (62.5%) experienced disease progression. The overall OS was roughly 6.8 months. The most frequently observed drug-related adverse events were grade 2/3/4 somnolence in 65%

In conclusion, with gradual dose escalation, thalidomide exhibited well-tolerated toxicity profile. While thalidomide demonstrated lower response rates, it offered disease stabilization in one-third of patients. Future studies should be targeted toward exploring different thali‐ domide analogs and doses as well as trial of combination therapy with other systemic management modalities. As of now, thalidomide use in the management of advanced HCC is

Single-agent irinotecan chemotherapy has been investigated in two phase II clinical trials for

In 2001, O'Reilly et al. (phase II) studied the role of irinotecan (starting dose of 125 mg once weekly for 4 weeks followed by a 2-week rest) in 14 patients with advanced unresectable HCC. The median number of irinotecan cycle administration was 1 (range: 1-6). Partial response was attained in only 1 patient (7%), which lasted for 7 weeks. Transient stable disease was attained in 1 patient (7%). Disease progression occurred in all the 12 remaining patients (86%). Signif‐ icant irinotecan-related adverse events were noted, mainly nausea, vomiting, diarrhea, fatigue,

In 2006, Boige et al. (multicenter phase II study) studied the role of irinotecan (dose was adjusted according to total bilirubin level) in 29 patients with advanced unresectable HCC. In total, 0, 1, and 12 patients experienced objective response, minor response, and disease stabilization, respectively. Median TTP was 3.1 months whereas the OS was 7.4 months. Grade 3/4 toxicities primarily compromised diarrhea (17%), anemia (24%), and neutropenia (47%).

grade 3 and grade 4 thalidomide-related adverse events, respectively.

114 Recent Advances in Liver Diseases and Surgery

whereas grade 3/4 reactions occurred in 20% of patients.

the management of patients with advanced unresectable HCC [62,.

not recommended.

*3.1.9. Irinotecan*

and neutropenia.

Various combinations of systemic cytotoxic chemotherapeutics have been investigated in patients with advanced HCC, such as cisplatin-based, gemcitabine-based, and oxaliplatinbased regimens.

Table 1 exhibits a summary of major phase I to II studies on combination systemic cytotoxic chemotherapy in patients with advanced inoperable HCC.

Overall, cisplatin-based combination chemotherapeutic schedules seem to yield greater objective response rates than non-cisplatin-based combination chemotherapeutic schedules. However, no single combination systemic chemotherapy regimen definitely appeared to offer superior or valuable survival advantages such as TTP, PFS, OS, and disease stabilization.

Regimens containing oxaliplatin plus short-term infusional 5-FU and leucovorin are most frequently utilized in the management of advanced colorectal cancer with hepatic metastases.

In 2013, Qin et al. (multicenter open-label, phase III randomized trial) examined the efficacy of single-agent doxorubicin (50 mg/m2 once every three weeks) versus modified FOLFOX4 regimen (infusional 5-fluorouracil, leucovorin, and oxaliplatin) in 371 Asian patients with primary locally advanced, inoperable, or metastatic HCC. Of note, 90 of all enrolled 371 patients (24.3%) had cirrhosis secondary to hepatitis B virus infection. In comparison with the doxorubicin group, the modified FOLFOX4 achieved slightly higher PFS (2.93 vs. 1.77 months, respectively), median OS (6.40 vs. 4.97 months, respectively), ORR (8.15%, vs. 2.67%, respec‐ tively), and DCR (52% vs. 32%, respectively). On continual follow-up, there was a statistically significant sustainable tendency toward improved OS with FOLFOX4 regimen versus doxorubicin (*P* = 0.04). Modified FOLFOX4-related adverse events were comparable to earlier studies. Both treated groups experienced similar grade 3/4 drug-related toxicities. The study concluded that the propensity toward enhanced OS, PFS, and ORR with modified FOLFOX4 regimen may offer some palliative advantages to the Asian HCC patients; however, a definite OS advantage cannot be deduced from their study, and further research was suggested.

#### **3.3. Interferon alpha (IFNα)**

Interferon alpha (IFNα) is an immunomodulatory cytokine (immunotherapy/biotherapy) that has exhibited antineoplastic effects against many neoplasms counting HCC.

#### *3.3.1. IFNα monotherapy*

As a minimum, three controlled trials have examined single-agent IFNα therapy in patients with far-advanced unresectable HCC; however, research outcomes were contradictory.


n: sample size; RR: response rate; DS: disease stabilization; TTP: time to progression; PFS: progression-free survival; OS: overall survival; NR: not reported; mon: months

**Table 1.** Summary of major phases I and II studies on combination systemic chemotherapy in patients with advanced inoperable HCC

In 1989, Lai et al. (Chinese prospective randomized trial) explored the efficacy of single-agent IFNα versus single-agent doxorubicin in 75 patients with advanced unresectable HCC. The IFNα group achieved a higher median OS than the doxorubicin group (8.3 months vs. 4.8 months), although it was not statistically significant. Doxorubicin-related adverse events included neutropenia and cardiotoxicity in approximately 25% of patients. Conversely, IFNαrelated adverse events included adrenal gland failure and dementia in roughly 3.8% of patients. Overall, IFNα achieved statistically significant robust cancer regression (*P* = 0.00199), less worsening cancers (*P* = 0.00017), less life-threatening long-lasting bone marrow suppres‐ sion (*P* = 0.01217), and less severe drug-related adverse events (*P* = 0.01383) when compared to doxorubicin group. The study concluded that IFNα was superior to doxorubicin in terms of cancer control as well as less lethal bone marrow suppression and adverse events.

In 1993, Lai et al. [66] (randomized controlled trial) examined the efficacy of IFNα (intramus‐ cular 50 × 106 IU/m2 3 times weekly) and no anticancer treatment in 35 and 36 advanced unresectable HCC Chinese patients, respectively. The IFNα group achieved a higher median OS than no anticancer group (14.5 vs. 7.5 months; *P* = 0.0471), as well as significant robust cancer regression (*P* < 0.0001) and less worsening (progressive) cancers (*P* = 0.001). Despite the IFNα dose was comparatively high, it was well tolerated; roughly 34% of patients had onethird to one-half dosage decreases as a result of continuous generalized weakness. Moreover, type 2 diabetes mellitus patients experienced mental worsening that could be related to IFNα treatment. The study concluded that IFNα was beneficial in a subset of Chinese patients with advanced unresectable HCC, in terms of cancer control (tumor regression) and extended disease-related survival expectancy.

However, the above-mentioned results of Lia et al. [66] were not validated and reciprocated in a second randomized clinical trial by Llovet et al. in 58 advanced HCC patients with ineligibility to undergo surgery, transplantation, or other treatment modalities. The study took place in year 2000 and randomized patients to receive either IFNα (*n* = 30) or BSC (*n* = 28). Of the 30 IFNα-treated patients, only 2 patients (6.6%) achieved objective partial responses. Although the 1-year and 2-year survival rates were higher in IFNα-treated vs. BSC groups (58% vs. 38% and 36% vs. 12%, respectively), there were no statistical significant differences. Although IFNα dose was greatly decreased, 23 (76.7%) of 30 patients experienced severe unbearable drug-related adverse events (toxicities) resulting in drug suspension in exactly 13 patients. The study concluded that IFNα was not appropriately endured by advanced HCC patients, and its administration did not yield beneficial advantages in the context of cancer progression and OS rates.

In conclusion, studies on single-agent IFNα therapy showed conflicting outcomes. Addition‐ ally, dose-related toxicities were frequent despite lower doses were administered. Clear-cut clinical benefits are uncertain and further research is needed.

#### *3.3.2. IFNα-based combination therapy*

There are two major IFN-based combination chemotherapeutic regimens: PIAF regimen and (5-FU plus IFNα) regimen.

#### *3.3.2.1. PIAF regimen*

In 1989, Lai et al. (Chinese prospective randomized trial) explored the efficacy of single-agent IFNα versus single-agent doxorubicin in 75 patients with advanced unresectable HCC. The IFNα group achieved a higher median OS than the doxorubicin group (8.3 months vs. 4.8 months), although it was not statistically significant. Doxorubicin-related adverse events included neutropenia and cardiotoxicity in approximately 25% of patients. Conversely, IFNαrelated adverse events included adrenal gland failure and dementia in roughly 3.8% of patients. Overall, IFNα achieved statistically significant robust cancer regression (*P* = 0.00199),

**Table 1.** Summary of major phases I and II studies on combination systemic chemotherapy in patients with advanced

n: sample size; RR: response rate; DS: disease stabilization; TTP: time to progression; PFS: progression-free survival; OS:

**Regimen Reference Authors Year Combination systemic**

116 Recent Advances in Liver Diseases and Surgery

Yang et al. 2004

Ikeda et al. 2005

Boucher et al.

Parikh et al.

Lombardi et al.

Louafi et al.

Zaanan et al.

overall survival; NR: not reported; mon: months

**Cisplatinbased regimen**

**Gemcitabi ne-based regimen**

**Oxaliplati n-based regimen**

inoperable HCC

[137]

[141]

[143]

[144]

[146]

**chemotherapy** *<sup>n</sup>* **RR (%)**

[134] Lee et al. 2004 Cisplatin plus doxorubicin 42 18.9 16.2 6.6 NR 7.3

[139] Shim et al. 2009 Cisplatin plus capecitabine 178 19.7 45 NR 2.8 10.5 [140] Lee et al. 2009 Cisplatin plus capecitabine 32 6.3 34.3 2 NR 12.2

[142] Chia et al. 2008 Gemcitabine and cisplatin 15 6.7 20 NR 1.5 4.5

2005 Gemcitabine and cisplatin 30 20 43 4.5 NR 5.3

Cisplatin, mitoxantrone, plus continuous infusion 5-

Cisplatin, mitoxantrone, plus continuous infusion 5-

<sup>2002</sup> Cisplatin, epirubicin plus infusional 5-FU

capecitabine

Gemcitabine plus pegylated liposomal doxorubicin

oxaliplatin (GEMOX)

oxaliplatin (XELOX)

<sup>2007</sup> Gemcitabine plus oxaliplatin (GEMOX)

<sup>2013</sup> Gemcitabine plus oxaliplatin (GEMOX)

FU

FU

[138] Park et al. <sup>2006</sup> Cisplatin, doxorubicin plus

2011

[145] Mir et al. <sup>2012</sup> Gemcitabine plus

[147] Boige et al. <sup>2007</sup> Gapecitabine plus

**DS (%)**

63 23.8 NR 2.5 NR 4.9

51 27 NR NR 4 11.6

21 14.5 NR 5.9 NR 10

29 24 20.7 3.7 NR 7.7

41 NR 24 NR 5.8 22.5

34 18 58 NR 6.3 11.5

18 18.8 18.8 NR 3.2 4.7

204 22 66 NR 4.5 11

50 6 72 NR 4.1 9.3

**TTP (mon)** **PFS (mon)** **OS (mon)**

> PIAF regimen is composed of cisplatin, IFNα, doxorubicin, and infusional 5-FU. PIAF regimen has been shown to exhibit active antitumor effects despite its significantly lethal drug-related toxic adverse events in patients with advanced HCC 8-. For example, in 1999, Leung et al. administered PIAF regimen in 50 patients. Around 13 patients (26%) experienced a partial response. The median OS was 8.9 months. The most frequent toxicities were mucositis and

myelosuppression. There were two events of drug-related mortality as a result of neutropenic sepsis.

In 2005, Yeo et al. (multinational randomized phase III study) examined the efficacy of singleagent doxorubicin (60 mg/m2 every three weeks) versus PIAF regimen (cisplatin: 20 mg/m2 on days 1-4; IFNα: 5 MU/m2 subcutaneously on days 1-4; doxorubicin: 40 mg/m2 on day 1; and 5- FU 400 mg/m2 on days 1-4) in 188 chemotherapy-naive patients with inoperable HCC. Although not statistically significant, the PIAF-treated group achieved higher ORR and median OS than the single-agent doxorubicin group (20.9% vs. 10.5% and 8.7 months vs. 6.8 months, respectively). However, as expected, drug-related adverse events were more notice‐ able and statistically significant in the PIAF-treated group than in doxorubicin-treated group, as follows: grade 3/4 hypokalemia (7% vs. 0%, respectively), grade 3/4 neutropenia (82% vs. 63%, respectively), and grade 3/4 thrombocytopenia (57% vs. 24%, respectively). The study concluded that although the PIAF-treated group achieved higher overall ORR and beneficial survival outcomes, the difference was statistically insignificant and not worthwhile. Addi‐ tionally, PIAF regimen incurred far greater statistically significant drug-related adverse events.

One potential clarification for the Yeo et al. study's failure to demonstrate a survival advantage may be attributed to the improper patient selection. Subsequently, the correlation significance between results of PIAF regimen and baseline liver function was exhibited in a retrospective analysis by Leung et al.. The study analyzed a series of roughly 150 patients with advanced inoperable HCC who received prior therapy with PIAF regimen. The study concluded that good risk patients (normal baseline total bilirubin levels and noncirrhotic liver) achieved higher statistically significant objective responses (50% vs. 6%) and prolonged survival rates than bad risk patients (total serum bilirubin level >0.6 mg/dL and cirrhotic liver) when medicated with systemic PIAF regimen.

In short, the role of PIAF chemotherapeutic schedule in the management of advanced inop‐ erable HCC remains unclear. Bearing in mind the lethal drug-related toxicity profile, it should be indicated only for physically and biochemically fit patients who possess appropriate performance status and minimal hepatic insufficiency.

#### *3.3.2.2. 5-FU plus IFNα*

Stuart et al. and Patt et al. had conflicting results. In 1996, Stuart et al. administered 5-FU (750 mg/m2 weekly) plus IFNα (9 MU three times weekly) in 10 patients with advanced HCC. The ORR and the OS were 0% and 10 months, respectively. It was concluded that the 5-FU plus IFNα regimen was not effective and drug-related toxicities were highly significant.

Moreover, in 2003, Patt et al. (phase II) administered 5-FU (200 mg/m2 /day for 3 weeks every 4-week interval) plus IFNα2b (4 million U/m2 for three times weekly) in 43 patients with advanced HCC. Liver cirrhosis was present among 71% of HCC. ORR was evaluable in only 28 patients, and it was 14% (all were partial responses). For all patients, the OS was 15.5 months. The study concluded that 5-FU plus IFNα is effective and can be tolerated by cirrhotic patients. Of note, several studies by Sakon et al., Ota et al., and Nagano et al. have examined the combination of systemic IFNα with intrahepatic arterial 5-FU in patients with primary advanced inoperable HCC complicated by major portal vein thrombosis. Interestingly, ORRs ranging from 33% to 73% were achieved. More specifically, chemotherapy responsiveness, TTP, and OS rates were higher in IFN-alpha type 2 receptor (IFNAR2)-positive HCC versus IFNAR2-negative HCC. It was concluded that chemotherapy responsiveness, TTP, and OS are significantly linked to expression of IFNAR2 in HCC patients receiving 5-FU plus IFNα combined chemotherapeutic regimen.

In conclusion, combinations of chemotherapeutics with interferon alpha (IFNα) seem to be active. However, definitive survival benefits are not clear.

#### **3.4. Conclusion**

myelosuppression. There were two events of drug-related mortality as a result of neutropenic

In 2005, Yeo et al. (multinational randomized phase III study) examined the efficacy of single-

days 1-4; IFNα: 5 MU/m2 subcutaneously on days 1-4; doxorubicin: 40 mg/m2 on day 1; and 5-

Although not statistically significant, the PIAF-treated group achieved higher ORR and median OS than the single-agent doxorubicin group (20.9% vs. 10.5% and 8.7 months vs. 6.8 months, respectively). However, as expected, drug-related adverse events were more notice‐ able and statistically significant in the PIAF-treated group than in doxorubicin-treated group, as follows: grade 3/4 hypokalemia (7% vs. 0%, respectively), grade 3/4 neutropenia (82% vs. 63%, respectively), and grade 3/4 thrombocytopenia (57% vs. 24%, respectively). The study concluded that although the PIAF-treated group achieved higher overall ORR and beneficial survival outcomes, the difference was statistically insignificant and not worthwhile. Addi‐ tionally, PIAF regimen incurred far greater statistically significant drug-related adverse

One potential clarification for the Yeo et al. study's failure to demonstrate a survival advantage may be attributed to the improper patient selection. Subsequently, the correlation significance between results of PIAF regimen and baseline liver function was exhibited in a retrospective analysis by Leung et al.. The study analyzed a series of roughly 150 patients with advanced inoperable HCC who received prior therapy with PIAF regimen. The study concluded that good risk patients (normal baseline total bilirubin levels and noncirrhotic liver) achieved higher statistically significant objective responses (50% vs. 6%) and prolonged survival rates than bad risk patients (total serum bilirubin level >0.6 mg/dL and cirrhotic liver) when

In short, the role of PIAF chemotherapeutic schedule in the management of advanced inop‐ erable HCC remains unclear. Bearing in mind the lethal drug-related toxicity profile, it should be indicated only for physically and biochemically fit patients who possess appropriate

Stuart et al. and Patt et al. had conflicting results. In 1996, Stuart et al. administered 5-FU (750

advanced HCC. Liver cirrhosis was present among 71% of HCC. ORR was evaluable in only 28 patients, and it was 14% (all were partial responses). For all patients, the OS was 15.5 months. The study concluded that 5-FU plus IFNα is effective and can be tolerated by cirrhotic patients.

IFNα regimen was not effective and drug-related toxicities were highly significant.

Moreover, in 2003, Patt et al. (phase II) administered 5-FU (200 mg/m2

 weekly) plus IFNα (9 MU three times weekly) in 10 patients with advanced HCC. The ORR and the OS were 0% and 10 months, respectively. It was concluded that the 5-FU plus

/day for 3 weeks every

for three times weekly) in 43 patients with

every three weeks) versus PIAF regimen (cisplatin: 20 mg/m2 on

on days 1-4) in 188 chemotherapy-naive patients with inoperable HCC.

sepsis.

events.

FU 400 mg/m2

agent doxorubicin (60 mg/m2

118 Recent Advances in Liver Diseases and Surgery

medicated with systemic PIAF regimen.

*3.3.2.2. 5-FU plus IFNα*

mg/m2

performance status and minimal hepatic insufficiency.

4-week interval) plus IFNα2b (4 million U/m2

The employment of systemic chemotherapy has been accompanied by low ORRs, no survival advantages, and high incidences of drug-related toxicities and adverse events. Moreover, there are no adequate data to endorse or approve any single-agent or combined chemotherapeutic regimens for the management of patients with advanced inoperable HCC [76].

Recently, chemotherapy is not being employed routinely for patients with advanced inoper‐ able HCC. This tendency can be attributed to three major rationales:


The arrival of novel molecularly targeted therapy (specifically sorafenib) is rapidly emerging as the standard of care in patients with advanced inoperable HCC.

That being said, systemic chemotherapy may still be regarded in patients whom their HCC get worse while on sorafenib and whom baseline liver function and performance status are adequate enough to endure it.

The chemotherapy-related adverse events of any single-agent or combined regimen should be deliberated cautiously in patients with progressive inoperable HCC, multiple comorbidities, and very short life expectancy. Generally speaking, systemic chemotherapy should be selectively administered to physically and medically fit patients who possess appropriate hepatic functional reserve. Moreover, such administration should be ideally considered only within the context of phase II and III clinical trials.

The choice of systemic chemotherapy should be guided by patients' functional hepatic reserve, physical fitness, prognosis, life expectancy, and most importantly availability of the best evidence-based medicine (randomized controlled phase III clinical trials).

Lastly, the reactivation of viral hepatitis may take place in HCC patients receiving aggressively exhaustive systemic chemotherapeutic regimens. Accordingly, it is crucial and greatly recommended to maintain antiviral therapies, whenever deemed necessarily.
