TACTIC: a multicentre, open-label, single-arm phase II trial of panitumumab, cisplatin, and gemcitabine in biliary tract cancer

PURPOSE

The phase II TACTIC trial prospectively selected patients with KRAS wild-type advanced biliary tract cancer for first-line treatment with panitumumab and combination chemotherapy.

METHODS

Of 78 patients screened, 85 % had KRAS wild-type tumours and 48 were enrolled. Participants received cisplatin 25 mg/m(2) and gemcitabine 1000 mg/m(2) on day 1 and day 8 of each 21-day cycle and panitumumab 9 mg/kg on day 1 of each cycle. Treatment was continued until disease progression, unacceptable toxicity, or request to discontinue. The primary endpoint was the clinical benefit rate (CBR) at 12 weeks (complete response, partial response, or stable disease). CBR of 70 % was considered to be of clinical interest. Secondary outcomes were progression-free survival, time to treatment failure, overall survival, CA19.9 response and safety.

RESULTS

Thirty-four patients had a clinical benefit at 12 weeks, an actuarial rate of 80 % (95 % CI 65-89 %). 46 % had a complete or partial response. Median progression-free survival was 8.0 months (95 % CI 5.1-9.9) and median overall survival 11.9 months (95 % CI 7.4-15.8). Infection accounted for 27 % of the grade 3 or 4 toxicity, with rash (13 %), fatigue (13 %), and hypomagnesemia (10 %) among the more common grade 3 or 4 non-haematological toxicities.

CONCLUSION

A marker-driven approach to patient selection was feasible in advanced biliary tract cancer in an Australian population. The combination of panitumumab, gemcitabine, and cisplatin in KRAS wild-type cancers was generally well tolerated and showed promising clinical efficacy. Further exploration of anti-EGFR therapy in a more selected population is warranted.

The mouse relaxin gene: nucleotide sequence and expression

Relaxin is a polypeptide hormone that has a variety of physiological effects both on remodelling of collagen and on uterine contractility. These are most apparent during pregnancy. The sequences of relaxin cDNAs derived from ovaries of late-pregnant random-bred Swiss mice have been established. Multiple subclones obtained from three independent polymerase chain reaction experiments were found to encode relaxins which were identical except at position 11 in the A chain (Ile or Val). All mouse relaxin cDNAs expressed in the ovary during pregnancy had an extra tyrosine inserted prior to the final A chain cysteine residue, a result confirmed by direct sequencing of relaxin peptides. Whilst this tyrosine insertion must have local effects on the folding of the A chain, structure-activity studies will clarify whether it perturbs functional interaction with the relaxin receptor. We have shown that there is a single relaxin gene in the mouse genome, and that expression during pregnancy occurs in the ovary but is not detectable in the placenta, uterus or fetus.

Structure of a genomic clone encoding biologically active human relaxin

Relaxin is a peptide hormone synthesized in the corpora lutea of ovaries during pregnancy and is released into the blood stream prior to parturition. Its major biological effect is to remodel the mammalian reproductive tract to facilitate the birth process. Determination of the structure of human relaxin is thus a first step in opening up the possibility of clinical intervention in cases of difficult labour. However, the limited availability of human ovaries during pregnancy has prevented both direct amino acid sequence determination and isolation of cDNA clones obtained from relaxin producing tissue. Our approach has therefore been to screen directly for a human relaxin gene using an homologous porcine relaxin cDNA probe. We report here the successful identification of a genomic clone from which the structure of the entire coding region of a human preprorelaxin gene has been determined. Synthesis of biologically active relaxin has shown that the novel gene structure described herein codes for an authentic human relaxin. We believe this is the first successful synthesis of a biologically active hormone whose structure was predicted solely from the structure of a genomic clone.

Porcine relaxin: molecular cloning and cDNA structure

Relaxin is a peptide hormone produced by the corpora lutea of ovaries during pregnancy, softening and lengthening the ligaments of the pelvis and softening the cervix in order to make childbirth easier. In attempts to determine the nucleotide sequence coding for relaxin, recombinant DNA techniques were used to obtain a cDNA clone bank from total mRNA isolated from the ovaries of pigs in late pregnancy. Clones were screened using cDNA initiated by synthetic oligonucleotide primers coding for the Trp Val Glu Ile sequence of the porcine relaxin B chain. The synthetic undecamer [5'-ATCTCCACCCA-3'] was found to prime a specific 32P-labeled cDNA of approximately 300 nucleotides containing B chain and signal peptide coding sequences, as verified by nucleic acid sequence analysis. This cDNA was used to probe the ovarian clone bank. Several clones containing large inserts which hybridized to this probe were subjected to sequence analysis and some of these were found to contain the preprorelaxin coding region, comprising a signal peptide of 24 amino acids, a B chain of 32 amino acids, a large C peptide of 104 amino acids, and an A chain of 22 amino acids. From the amino acid sequence of prorelaxin derived in this way, it appears that the processing of prorelaxin involves two enzymes with chymotrypsin-like and trypsin-like specificity, respectively. In comparisons of porcine and rat preprorelaxins, the C region had as much amino acid sequence homology as the B and A chains. The C region is also rich in charged amino acids, suggesting a role for it beyond simply ensuring proper disulfide bond formation.

Molecular cloning and characterization of cDNA sequences coding for rat relaxin

Using a synthetic oligonucleotide primer, cloned DNA fragments, each containing the entire coding sequence of rat relaxin, have been isolated from a clone 'bank' of ovarian mRNA sequences. The nucleotide sequence of these clones demonstrates that relaxin is synthesized as a preprorelaxin molecule with an unexpectedly large connecting peptide of 105 amino acid residues.